Predicting patient outcomes early in the admission process is an important part of hospital management [1,2]. Estimating a patient’s length of stay, for instance, assists in logistics such as allocation of hospital resources [2]. The ability to predict patient-specific healthcare costs helps us gain understanding of the situations that lead to high healthcare costs and can inform future decisions in preventative care, which has been shown to be one of the best ways to decrease healthcare cost [3].
For our final project we will be using the Statewide Planning and Research Cooperative System (SPARCS) Inpatient De-identified File dataset (originally provided the New York State Department of Health [4]) as it appears on Kaggle to predict length of stay (LOS) in days, a metric used to assess hospital efficiency and quality of care [2,5,6], and patient cost per day. The full dataset contains nearly 2.35 million de-identified observations and 37 features about hospital inpatient discharge data for 2015.
report_table <- function(x, w = NULL, h = "500px", ec = NULL, rownames = FALSE) {
scroll_box(kable_paper(kable_styling(kbl(x, row.names = rownames), full_width = T,
position = "center", bootstrap_options = c("striped", "hover"), fixed_thead = T),
full_width = T), fixed_thead = T, height = h, width = w, extra_css = ec)
}
report_plot <- function(x) {
girafe_options(girafe(code = print(x)), opts_sizing(rescale = TRUE), opts_tooltip(opacity = 0.9,
css = "background-color:gray;color:white;padding:2px;border-radius:2px;"),
opts_hover(css = "cursor:pointer;"))
}# Import the data
dat <- read.csv("input/Hospital_Inpatient_Discharges__SPARCS_De-Identified___2015.csv",
colClasses = c(rep("factor", 10), "character", rep("factor", 21), "numeric",
rep("factor", 2), rep("character", 2)), na.strings = c("NA", ""))In addition to the columns found in the Kaggle dataset, we will also be using the New York State Department of Health’s “Health Facility General Information” dataset [7] in order to incorporate Facility Type and Ownership Type variables into our analysis by matching on Facility ID.
# read in facility id and description columns from facility information dataset
dat_hf <- read.csv("input/Health_Facility_General_Information.csv", colClasses = c("factor",
rep("NULL", 2), "factor", rep("NULL", 28), "factor", rep("NULL", 3)), skipNul = TRUE)
# remove duplicate rows
dat_hf <- dat_hf[!duplicated(dat_hf), ]
# rename columns
colnames(dat_hf) <- c("Facility.Id", "Facility.Type", "Ownership.Type")
# merge datasets by Facility Id
dat <- droplevels(merge(dat, dat_hf, by = "Facility.Id", all.x = TRUE, sort = FALSE))| Facility.Id | Health.Service.Area | Hospital.County | Operating.Certificate.Number | Facility.Name | Age.Group | Zip.Code…3.digits | Gender | Race | Ethnicity | Length.of.Stay | Type.of.Admission | Patient.Disposition | Discharge.Year | CCS.Diagnosis.Code | CCS.Diagnosis.Description | CCS.Procedure.Code | CCS.Procedure.Description | APR.DRG.Code | APR.DRG.Description | APR.MDC.Code | APR.MDC.Description | APR.Severity.of.Illness.Code | APR.Severity.of.Illness.Description | APR.Risk.of.Mortality | APR.Medical.Surgical.Description | Payment.Typology.1 | Payment.Typology.2 | Payment.Typology.3 | Attending.Provider.License.Number | Operating.Provider.License.Number | Other.Provider.License.Number | Birth.Weight | Abortion.Edit.Indicator | Emergency.Department.Indicator | Total.Charges | Total.Costs | Facility.Type | Ownership.Type |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 30 to 49 | 147 | F | White | Not Span/Hispanic | 5 | Emergency | Home or Self Care | 2015 | 127 | Chronic obstructive pulmonary disease and bronchiectasis | 0 | NO PROC | 140 | Chronic obstructive pulmonary disease | 4 | Diseases and Disorders of the Respiratory System | 3 | Major | Moderate | Medical | Medicare | Medicaid | NA | 197453 | NA | NA | 0 | N | Y | $18309.20 | $10309.98 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | M | White | Not Span/Hispanic | 3 | Emergency | Home w/ Home Health Services | 2015 | 108 | Congestive heart failure; nonhypertensive | 58 | HEMODIALYSIS | 194 | Heart failure | 5 | Diseases and Disorders of the Circulatory System | 3 | Major | Major | Medical | Medicare | NA | NA | 40003993 | 193890 | NA | 0 | N | Y | $12310.76 | $6034.20 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 148 | F | White | Not Span/Hispanic | 9 | Emergency | Home or Self Care | 2015 | 2 | Septicemia (except in labor) | 222 | BLOOD TRANSFUSION | 720 | Septicemia & disseminated infections | 18 | Infectious and Parasitic Diseases, Systemic or Unspecified Sites | 4 | Extreme | Major | Medical | Medicare | NA | NA | 266409 | 266335 | NA | 0 | N | Y | $48242.20 | $20494.93 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | OOS | F | White | Not Span/Hispanic | 2 | Elective | Home or Self Care | 2015 | 100 | Acute myocardial infarction | 47 | DX CARDIAC CATHETERIZTN | 190 | Acute myocardial infarction | 5 | Diseases and Disorders of the Circulatory System | 2 | Moderate | Moderate | Medical | Medicaid | NA | NA | 90271326 | 183914 | NA | 0 | N | N | $12794.77 | $7369.54 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 4 | Emergency | Home w/ Home Health Services | 2015 | 100 | Acute myocardial infarction | 0 | NO PROC | 190 | Acute myocardial infarction | 5 | Diseases and Disorders of the Circulatory System | 2 | Moderate | Moderate | Medical | Medicare | NA | NA | 266141 | NA | NA | 0 | N | Y | $10292.68 | $4001.00 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 18 to 29 | 141 | M | White | Not Span/Hispanic | 7 | Emergency | Short-term Hospital | 2015 | 130 | Pleurisy; pneumothorax; pulmonary collapse | 39 | INCISION OF PLEURA | 143 | Other respiratory diagnoses except signs, symptoms & minor diagnoses | 4 | Diseases and Disorders of the Respiratory System | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 271326 | 273279 | NA | 0 | N | Y | $14177.09 | $7711.28 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 141 | M | White | Not Span/Hispanic | 9 | Emergency | Home w/ Home Health Services | 2015 | 660 | Alcohol-related disorders | 231 | OTHER THERAPEUTIC PRCS | 280 | Alcoholic liver disease | 7 | Diseases and Disorders of the Hepatobiliary System and Pancreas | 3 | Major | Major | Medical | Medicare | NA | NA | 40003993 | 239912 | NA | 0 | N | Y | $21448.73 | $9511.37 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 0 to 17 | 147 | F | White | Not Span/Hispanic | 3 | Newborn | Home or Self Care | 2015 | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 256322 | NA | NA | 3400 | N | N | $2393.10 | $1327.02 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 0 to 17 | 147 | F | White | Not Span/Hispanic | 4 | Newborn | Home or Self Care | 2015 | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 2 | Moderate | Minor | Medical | Medicaid | NA | NA | 182357 | NA | NA | 2900 | N | N | $4191.00 | $2136.17 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 4 | Elective | Skilled Nursing Home | 2015 | 204 | Other non-traumatic joint disorders | 152 | ARTHROPLASTY KNEE | 302 | Knee joint replacement | 8 | Diseases and Disorders of the Musculoskeletal System and Conn Tissue | 1 | Minor | Minor | Surgical | Medicare | Private Health Insurance | NA | 257934 | 257934 | NA | 0 | N | N | $29514.01 | $59253.25 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 0 to 17 | 147 | F | White | Not Span/Hispanic | 1 | Newborn | Home or Self Care | 2015 | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 231077 | NA | NA | 2700 | N | N | $813.50 | $459.49 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 140 | M | White | Not Span/Hispanic | 4 | Emergency | Home or Self Care | 2015 | 2 | Septicemia (except in labor) | 0 | NO PROC | 720 | Septicemia & disseminated infections | 18 | Infectious and Parasitic Diseases, Systemic or Unspecified Sites | 3 | Major | Major | Medical | Medicare | Blue Cross/Blue Shield | NA | 271326 | NA | NA | 0 | N | Y | $10114.85 | $4943.62 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 0 to 17 | 147 | M | White | Not Span/Hispanic | 2 | Newborn | Home or Self Care | 2015 | 218 | Liveborn | 115 | CIRCUMCISION | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 231077 | 231077 | NA | 2600 | N | N | $1510.00 | $929.25 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | M | White | Not Span/Hispanic | 2 | Emergency | Home or Self Care | 2015 | 129 | Aspiration pneumonitis; food/vomitus | 0 | NO PROC | 137 | Major respiratory infections & inflammations | 4 | Diseases and Disorders of the Respiratory System | 3 | Major | Moderate | Medical | Medicaid | NA | NA | 266141 | NA | NA | 0 | N | Y | $7330.35 | $2502.54 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 0 to 17 | 147 | F | White | Not Span/Hispanic | 2 | Newborn | Home or Self Care | 2015 | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 2 | Moderate | Minor | Medical | Medicaid | NA | NA | 237701 | NA | NA | 2900 | N | N | $1528.50 | $907.69 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 0 to 17 | 148 | M | White | Not Span/Hispanic | 1 | Newborn | Home or Self Care | 2015 | 218 | Liveborn | 115 | CIRCUMCISION | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Medicaid | Medicaid | NA | 237701 | 60001091 | NA | 3200 | N | N | $851.50 | $496.77 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 18 to 29 | 147 | F | White | Not Span/Hispanic | 3 | Elective | Home or Self Care | 2015 | 187 | Malposition; malpresentation | 134 | CESAREAN SECTION | 540 | Cesarean delivery | 14 | Pregnancy, Childbirth and the Puerperium | 1 | Minor | Minor | Surgical | Medicaid | NA | NA | 147483 | 147483 | NA | 0 | N | N | $8550.41 | $4645.93 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 10 | Emergency | Inpatient Rehabilitation Facility | 2015 | 51 | Other endocrine disorders | 0 | NO PROC | 424 | Other endocrine disorders | 10 | Endocrine, Nutritional and Metabolic Diseases and Disorders | 3 | Major | Major | Medical | Medicare | NA | NA | 40003993 | NA | NA | 0 | N | Y | $17076.26 | $9286.99 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 13 | Emergency | Skilled Nursing Home | 2015 | 2 | Septicemia (except in labor) | 54 | OT VASC CATH; NOT HEART | 720 | Septicemia & disseminated infections | 18 | Infectious and Parasitic Diseases, Systemic or Unspecified Sites | 4 | Extreme | Extreme | Medical | Medicare | Private Health Insurance | NA | 197563 | 258490 | NA | 0 | N | Y | $58394.67 | $40505.88 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | OOS | F | White | Not Span/Hispanic | 7 | Emergency | Home w/ Home Health Services | 2015 | 127 | Chronic obstructive pulmonary disease and bronchiectasis | 0 | NO PROC | 140 | Chronic obstructive pulmonary disease | 4 | Diseases and Disorders of the Respiratory System | 2 | Moderate | Moderate | Medical | Medicare | Federal/State/Local/VA | NA | 271326 | NA | NA | 0 | N | Y | $17623.18 | $8901.52 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | M | White | Not Span/Hispanic | 2 | Emergency | Home or Self Care | 2015 | 103 | Pulmonary heart disease | 0 | NO PROC | 134 | Pulmonary embolism | 4 | Diseases and Disorders of the Respiratory System | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 271326 | NA | NA | 0 | N | Y | $8016.13 | $2608.73 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 30 to 49 | 147 | F | White | Not Span/Hispanic | 7 | Emergency | Home or Self Care | 2015 | 657 | Mood disorders | 0 | NO PROC | 753 | Bipolar disorders | 19 | Mental Diseases and Disorders | 2 | Moderate | Minor | Medical | Medicaid | NA | NA | 40003856 | NA | NA | 0 | N | Y | $7380.68 | $9880.01 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 0 to 17 | 147 | M | White | Not Span/Hispanic | 3 | Newborn | Home or Self Care | 2015 | 218 | Liveborn | 115 | CIRCUMCISION | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 182357 | 182357 | NA | 3200 | N | N | $1964.00 | $1261.62 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | M | White | Not Span/Hispanic | 4 | Emergency | Home w/ Home Health Services | 2015 | 11 | Cancer of head and neck | 71 | GASTROSTOMY; TEMP/PERM | 110 | Ear, nose, mouth, throat, cranial/facial malignancies | 3 | Ear, Nose, Mouth, Throat and Craniofacial Diseases and Disorders | 3 | Major | Moderate | Medical | Private Health Insurance | Federal/State/Local/VA | NA | 256322 | 251671 | NA | 0 | N | Y | $7568.00 | $5248.66 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 148 | M | White | Not Span/Hispanic | 6 | Emergency | Skilled Nursing Home | 2015 | 50 | Diabetes mellitus with complications | 0 | NO PROC | 468 | Other kidney & urinary tract diagnoses, signs & symptoms | 11 | Diseases and Disorders of the Kidney and Urinary Tract | 3 | Major | Major | Medical | Medicare | Private Health Insurance | NA | 276511 | NA | NA | 0 | N | Y | $10902.98 | $5216.32 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 4 | Emergency | Home or Self Care | 2015 | 127 | Chronic obstructive pulmonary disease and bronchiectasis | 216 | RESP INTUB/MECH VENTIL | 140 | Chronic obstructive pulmonary disease | 4 | Diseases and Disorders of the Respiratory System | 3 | Major | Extreme | Medical | Medicare | NA | NA | 165169 | 273279 | NA | 0 | N | Y | $14544.13 | $6380.94 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 0 to 17 | 141 | F | White | Not Span/Hispanic | 2 | Newborn | Home or Self Care | 2015 | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 182357 | NA | NA | 2500 | N | N | $1216.42 | $811.52 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 30 to 49 | 148 | F | White | Not Span/Hispanic | 2 | Emergency | Home or Self Care | 2015 | 657 | Mood disorders | 0 | NO PROC | 753 | Bipolar disorders | 19 | Mental Diseases and Disorders | 1 | Minor | Minor | Medical | Federal/State/Local/VA | NA | NA | 267162 | NA | NA | 0 | N | Y | $2415.05 | $1876.09 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 30 to 49 | 147 | F | White | Not Span/Hispanic | 1 | Elective | Home or Self Care | 2015 | 196 | Other pregnancy and delivery including normal | 137 | OT PRCS TO ASSIST DELIV | 541 | Vaginal delivery w sterilization &/or D&C | 14 | Pregnancy, Childbirth and the Puerperium | 1 | Minor | Minor | Surgical | Federal/State/Local/VA | NA | NA | 197563 | 197563 | NA | 0 | N | N | $4256.44 | $3224.37 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 18 to 29 | 147 | M | White | Not Span/Hispanic | 5 | Emergency | Home or Self Care | 2015 | 657 | Mood disorders | 0 | NO PROC | 751 | Major depressive disorders & other/unspecified psychoses | 19 | Mental Diseases and Disorders | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | Medicaid | NA | 267162 | NA | NA | 0 | N | Y | $5176.78 | $6235.63 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 9 | Emergency | Skilled Nursing Home | 2015 | 42 | Secondary malignancies | 231 | OTHER THERAPEUTIC PRCS | 952 | Nonextensive procedure unrelated to principal diagnosis | 4 | Diseases and Disorders of the Respiratory System | 3 | Major | Major | Surgical | Medicare | Medicaid | NA | 271326 | 239912 | NA | 0 | N | Y | $22783.03 | $12623.80 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 2 | Emergency | Home w/ Home Health Services | 2015 | 127 | Chronic obstructive pulmonary disease and bronchiectasis | 0 | NO PROC | 140 | Chronic obstructive pulmonary disease | 4 | Diseases and Disorders of the Respiratory System | 2 | Moderate | Moderate | Medical | Medicare | NA | NA | 271326 | NA | NA | 0 | N | Y | $15927.38 | $5406.24 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | M | White | Not Span/Hispanic | 3 | Emergency | Home or Self Care | 2015 | 62 | Coagulation and hemorrhagic disorders | 76 | COLONOSCOPY AND BIOPSY | 661 | Coagulation & platelet disorders | 16 | Diseases and Disorders of Blood, Blood Forming Organs and Immunological Disorders | 3 | Major | Moderate | Medical | Medicare | NA | NA | 40003993 | 137074 | NA | 0 | N | Y | $12730.75 | $4252.01 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 6 | Emergency | Inpatient Rehabilitation Facility | 2015 | 135 | Intestinal infection | 222 | BLOOD TRANSFUSION | 248 | Major gastrointestinal & peritoneal infections | 6 | Diseases and Disorders of the Digestive System | 4 | Extreme | Major | Medical | Medicare | Medicaid | NA | 266141 | 251437 | NA | 0 | N | Y | $17490.31 | $9772.47 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 10 | Emergency | Skilled Nursing Home | 2015 | 157 | Acute and unspecified renal failure | 222 | BLOOD TRANSFUSION | 460 | RENAL FAILURE | 11 | Diseases and Disorders of the Kidney and Urinary Tract | 4 | Extreme | Extreme | Medical | Medicare | Private Health Insurance | NA | 266141 | 276511 | NA | 0 | N | Y | $19586.28 | $8822.05 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 18 to 29 | 147 | F | White | Not Span/Hispanic | 1 | Emergency | Left Against Medical Advice | 2015 | 196 | Other pregnancy and delivery including normal | 0 | NO PROC | 561 | Postpartum & post abortion diagnoses w/o procedure | 14 | Pregnancy, Childbirth and the Puerperium | 1 | Minor | Minor | Medical | Private Health Insurance | NA | NA | 256322 | NA | NA | 0 | N | Y | $2203.28 | $894.46 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 140 | F | White | Not Span/Hispanic | 6 | Emergency | Skilled Nursing Home | 2015 | 159 | Urinary tract infections | 0 | NO PROC | 463 | Kidney & urinary tract infections | 11 | Diseases and Disorders of the Kidney and Urinary Tract | 3 | Major | Major | Medical | Medicare | NA | NA | 170599 | NA | NA | 0 | N | Y | $10387.40 | $4716.31 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 2 | Elective | Home or Self Care | 2015 | 44 | Neoplasms of unspecified nature or uncertain behavior | 124 | HYSTERECTOMY; AB/VAG | 513 | Uterine & adnexa procedures for non-malignancy except leiomyoma | 13 | Diseases and Disorders of the Female Reproductive System | 1 | Minor | Minor | Surgical | Blue Cross/Blue Shield | NA | NA | 155651 | 155651 | NA | 0 | N | N | $15570.16 | $12042.57 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 6 | Emergency | Home w/ Home Health Services | 2015 | 159 | Urinary tract infections | 54 | OT VASC CATH; NOT HEART | 463 | Kidney & urinary tract infections | 11 | Diseases and Disorders of the Kidney and Urinary Tract | 2 | Moderate | Major | Medical | Medicare | NA | NA | 161469 | 161469 | NA | 0 | N | Y | $17509.05 | $9587.57 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 30 to 49 | 147 | F | White | Not Span/Hispanic | 2 | Emergency | Inpatient Rehabilitation Facility | 2015 | 3 | Bacterial infection; unspecified site | 54 | OT VASC CATH; NOT HEART | 724 | Other infectious & parasitic diseases | 18 | Infectious and Parasitic Diseases, Systemic or Unspecified Sites | 3 | Major | Minor | Medical | Medicaid | NA | NA | 197563 | 197563 | NA | 0 | N | Y | $12811.88 | $6991.10 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 6 | Elective | Home w/ Home Health Services | 2015 | 197 | Skin and subcutaneous tissue infections | 0 | NO PROC | 383 | Cellulitis & other skin infections | 9 | Diseases and Disorders of the Skin, Subcutaneous Tissue and Breast | 2 | Moderate | Minor | Medical | Medicare | NA | NA | 161469 | NA | NA | 0 | N | N | $10372.37 | $5325.48 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 141 | F | White | Not Span/Hispanic | 3 | Emergency | Home or Self Care | 2015 | 657 | Mood disorders | 0 | NO PROC | 751 | Major depressive disorders & other/unspecified psychoses | 19 | Mental Diseases and Disorders | 2 | Moderate | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 275408 | NA | NA | 0 | N | Y | $3688.28 | $2800.36 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 4 | Emergency | Skilled Nursing Home | 2015 | 153 | Gastrointestinal hemorrhage | 92 | OTHER BOWEL DX PRCS | 253 | Other & unspecified gastrointestinal hemorrhage | 6 | Diseases and Disorders of the Digestive System | 3 | Major | Extreme | Medical | Medicare | Medicaid | NA | 141658 | 251671 | NA | 0 | N | Y | $11755.70 | $4987.86 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 148 | F | White | Not Span/Hispanic | 6 | Emergency | Hosp Basd Medicare Approved Swing Bed | 2015 | 197 | Skin and subcutaneous tissue infections | 54 | OT VASC CATH; NOT HEART | 383 | Cellulitis & other skin infections | 9 | Diseases and Disorders of the Skin, Subcutaneous Tissue and Breast | 1 | Minor | Moderate | Medical | Blue Cross/Blue Shield | NA | NA | 266141 | 252850 | NA | 0 | N | Y | $14615.33 | $8883.59 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 2 | Emergency | Home or Self Care | 2015 | 153 | Gastrointestinal hemorrhage | 92 | OTHER BOWEL DX PRCS | 253 | Other & unspecified gastrointestinal hemorrhage | 6 | Diseases and Disorders of the Digestive System | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 256322 | 137074 | NA | 0 | N | Y | $7648.80 | $2457.26 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 18 to 29 | 147 | F | White | Not Span/Hispanic | 3 | Elective | Home or Self Care | 2015 | 184 | Early or threatened labor | 137 | OT PRCS TO ASSIST DELIV | 560 | Vaginal delivery | 14 | Pregnancy, Childbirth and the Puerperium | 2 | Moderate | Minor | Medical | Medicaid | NA | NA | 60420192 | 60420192 | NA | 0 | N | N | $5429.70 | $3396.63 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | M | White | Not Span/Hispanic | 2 | Emergency | Home or Self Care | 2015 | 146 | Diverticulosis and diverticulitis | 0 | NO PROC | 244 | Diverticulitis & diverticulosis | 6 | Diseases and Disorders of the Digestive System | 2 | Moderate | Moderate | Medical | Medicare | Blue Cross/Blue Shield | Private Health Insurance | 197453 | NA | NA | 0 | N | Y | $4622.00 | $1691.56 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | OOS | M | White | Not Span/Hispanic | 2 | Elective | Short-term Hospital | 2015 | 100 | Acute myocardial infarction | 47 | DX CARDIAC CATHETERIZTN | 190 | Acute myocardial infarction | 5 | Diseases and Disorders of the Circulatory System | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 271326 | 183914 | NA | 0 | N | N | $22161.20 | $12458.00 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 50 to 69 | 147 | M | White | Not Span/Hispanic | 3 | Emergency | Home w/ Home Health Services | 2015 | 226 | Fracture of neck of femur (hip) | 153 | HIP REPLACEMENT,TOT/PRT | 301 | Hip joint replacement | 8 | Diseases and Disorders of the Musculoskeletal System and Conn Tissue | 1 | Minor | Minor | Surgical | Blue Cross/Blue Shield | Private Health Insurance | NA | 172880 | 172880 | 90012432 | 0 | N | Y | $30155.82 | $55346.46 | Hospital | Not for Profit Corporation |
| 66 | Western NY | Cattaraugus | 0401001 | Olean General Hospital | 70 or Older | 147 | M | White | Not Span/Hispanic | 4 | Emergency | Skilled Nursing Home | 2015 | 149 | Biliary tract disease | 0 | NO PROC | 284 | Disorders of gallbladder & biliary tract | 7 | Diseases and Disorders of the Hepatobiliary System and Pancreas | 3 | Major | Major | Medical | Medicare | Medicaid | NA | 165169 | NA | NA | 0 | N | Y | $13651.03 | $5599.37 | Hospital | Not for Profit Corporation |
# rename Zip.Code...3.digits
colnames(dat)[which(colnames(dat) == "Zip.Code...3.digits")] <- "Zip"
# make Length.of.Stay numeric
dat[which(dat$Length.of.Stay == "120 +"), "Length.of.Stay"] <- "121"
dat$Length.of.Stay <- as.numeric(dat$Length.of.Stay)
# Remove dollar sign from Total Costs and Total Charges and make numeric
dat$Total.Charges <- as.numeric(substring(dat$Total.Charges, 2))
dat$Total.Costs <- as.numeric(substring(dat$Total.Costs, 2))
# Create cost per day variable
dat$Cost.per.Day <- dat$Total.Costs/dat$Length.of.Stay
# Re-order factors
dat$APR.MDC.Code <- factor(dat$APR.MDC.Code, levels = as.character(0:25))
dat$APR.Severity.of.Illness.Description <- factor(dat$APR.Severity.of.Illness.Description,
levels = c("Minor", "Moderate", "Major", "Extreme"))
# Remove columns: discharge year is 2015 for all rows provider license numbers
# are many-leveled factors unused in analysis patient disposition information is
# known at end of hospital stay, not at admission
dat <- dat[, which(!(colnames(dat) %in% c("Discharge.Year", "Operating.Certificate.Number",
"Attending.Provider.License.Number", "Operating.Provider.License.Number", "Other.Provider.License.Number",
"Patient.Disposition")))]## Facility.Id Health.Service.Area Hospital.County
## 1456 : 55896 New York City :1092224 Manhattan: 400991
## 541 : 47712 Long Island : 339277 Kings : 248955
## 1464 : 47518 Hudson Valley : 245945 Queens : 197202
## 1458 : 42997 Capital/Adiron: 167367 Bronx : 188160
## 1169 : 42720 Western NY : 163927 Nassau : 181164
## (Other):2107006 (Other) : 335109 (Other) :1127377
## NA's : 2911 NA's : 2911 NA's : 2911
## Facility.Name
## Mount Sinai Hospital : 55896
## North Shore University Hospital : 47712
## New York Presbyterian Hospital - Columbia Presbyterian Center : 47518
## New York Presbyterian Hospital - New York Weill Cornell Center: 42997
## Montefiore Medical Center - Henry & Lucy Moses Div : 42720
## Maimonides Medical Center : 41909
## (Other) :2068008
## Age.Group Zip Gender
## 0 to 17 :352588 112 : 313548 F:1307507
## 18 to 29 :247223 104 : 224997 M:1039206
## 30 to 49 :458457 117 : 182030 U: 47
## 50 to 69 :645004 100 : 170127
## 70 or Older:643488 113 : 121512
## (Other):1330119
## NA's : 4427
## Race Ethnicity Length.of.Stay
## Black/African American: 444944 Multi-ethnic : 8693 Min. : 1.000
## Multi-racial : 22380 Not Span/Hispanic:1954468 1st Qu.: 2.000
## Other Race : 544559 Spanish/Hispanic : 278587 Median : 3.000
## White :1334877 Unknown : 105012 Mean : 5.476
## 3rd Qu.: 6.000
## Max. :121.000
##
## Type.of.Admission CCS.Diagnosis.Code
## Elective : 447280 218 : 226813
## Emergency :1488031 2 : 108709
## Newborn : 227100 203 : 63476
## Not Available: 1173 108 : 58760
## Trauma : 6394 657 : 56950
## Urgent : 176782 122 : 47082
## (Other):1784970
## CCS.Diagnosis.Description
## Liveborn : 226813
## Septicemia (except in labor) : 108709
## Osteoarthritis : 63476
## Congestive heart failure; nonhypertensive : 58760
## Mood disorders : 56950
## Pneumonia (except that caused by tuberculosis or sexually transmitted disease): 47082
## (Other) :1784970
## CCS.Procedure.Code CCS.Procedure.Description APR.DRG.Code
## 0 : 610252 NO PROC : 610252 640 : 198114
## 231 : 175510 OTHER THERAPEUTIC PRCS : 175510 560 : 147035
## 137 : 83658 OT PRCS TO ASSIST DELIV: 83658 720 : 94966
## 228 : 81060 PROPHYLACTIC VAC/INOCUL: 81060 540 : 76059
## 134 : 74806 CESAREAN SECTION : 74806 194 : 56375
## 216 : 74069 RESP INTUB/MECH VENTIL : 74069 139 : 43040
## (Other):1247405 (Other) :1247405 (Other):1731171
## APR.DRG.Description
## Neonate birthwt >2499g, normal newborn or neonate w other problem: 198114
## Vaginal delivery : 147035
## Septicemia & disseminated infections : 94966
## Cesarean delivery : 76059
## Heart failure : 56375
## Other pneumonia : 43040
## (Other) :1731171
## APR.MDC.Code
## 5 :289531
## 14 :254079
## 15 :231854
## 8 :204675
## 6 :197654
## 4 :196486
## (Other):972481
## APR.MDC.Description
## Diseases and Disorders of the Circulatory System :289531
## Pregnancy, Childbirth and the Puerperium :254079
## Newborns and Other Neonates with Conditions Originating in the Perinatal Period:231854
## Diseases and Disorders of the Musculoskeletal System and Conn Tissue :204675
## Diseases and Disorders of the Digestive System :197654
## Diseases and Disorders of the Respiratory System :196486
## (Other) :972481
## APR.Severity.of.Illness.Code APR.Severity.of.Illness.Description
## 0: 112 Minor :785526
## 1:785526 Moderate:897251
## 2:897251 Major :517129
## 3:517129 Extreme :146742
## 4:146742 NA's : 112
##
##
## APR.Risk.of.Mortality APR.Medical.Surgical.Description
## Extreme : 121749 Medical :1779501
## Major : 335242 Not Applicable: 112
## Minor :1389588 Surgical : 567147
## Moderate: 500069
## NA's : 112
##
##
## Payment.Typology.1 Payment.Typology.2
## Medicare :876258 Medicaid :554720
## Medicaid :717285 Self-Pay :366081
## Private Health Insurance:334558 Medicare :336787
## Blue Cross/Blue Shield :275552 Private Health Insurance:161646
## Self-Pay : 67879 Blue Cross/Blue Shield :124393
## Miscellaneous/Other : 28985 (Other) : 40787
## (Other) : 46243 NA's :762346
## Payment.Typology.3 Birth.Weight Abortion.Edit.Indicator
## Self-Pay : 479620 Min. : 0.0 N:2343849
## Medicaid : 124629 1st Qu.: 0.0 Y: 2911
## Private Health Insurance: 34861 Median : 0.0
## Blue Cross/Blue Shield : 21473 Mean : 326.4
## Medicare : 18971 3rd Qu.: 0.0
## (Other) : 21636 Max. :9900.0
## NA's :1645570
## Emergency.Department.Indicator Total.Charges Total.Costs
## N: 977695 Min. : 0 Min. : 0
## Y:1369065 1st Qu.: 12027 1st Qu.: 4724
## Median : 23481 Median : 8791
## Mean : 43206 Mean : 15985
## 3rd Qu.: 46607 3rd Qu.: 16835
## Max. :7248391 Max. :5236615
##
## Facility.Type
## Hospital :2296627
## Primary Care Hospital - Critical Access Hospital: 9361
## NA's : 40772
##
##
##
##
## Ownership.Type Cost.per.Day
## County : 3711 Min. : 0
## Municipality : 186339 1st Qu.: 1611
## Not for Profit Corporation:1979711 Median : 2465
## Public Benefit Corporation: 47937 Mean : 3482
## State : 88290 3rd Qu.: 4098
## NA's : 40772 Max. :1017571
##
| Facility.Id | Health.Service.Area | Hospital.County | Facility.Name | Age.Group | Zip | Gender | Race | Ethnicity | Length.of.Stay | Type.of.Admission | CCS.Diagnosis.Code | CCS.Diagnosis.Description | CCS.Procedure.Code | CCS.Procedure.Description | APR.DRG.Code | APR.DRG.Description | APR.MDC.Code | APR.MDC.Description | APR.Severity.of.Illness.Code | APR.Severity.of.Illness.Description | APR.Risk.of.Mortality | APR.Medical.Surgical.Description | Payment.Typology.1 | Payment.Typology.2 | Payment.Typology.3 | Birth.Weight | Abortion.Edit.Indicator | Emergency.Department.Indicator | Total.Charges | Total.Costs | Facility.Type | Ownership.Type | Cost.per.Day |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 66 | Western NY | Cattaraugus | Olean General Hospital | 30 to 49 | 147 | F | White | Not Span/Hispanic | 5 | Emergency | 127 | Chronic obstructive pulmonary disease and bronchiectasis | 0 | NO PROC | 140 | Chronic obstructive pulmonary disease | 4 | Diseases and Disorders of the Respiratory System | 3 | Major | Moderate | Medical | Medicare | Medicaid | NA | 0 | N | Y | 18309.20 | 10309.98 | Hospital | Not for Profit Corporation | 2061.9960 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | M | White | Not Span/Hispanic | 3 | Emergency | 108 | Congestive heart failure; nonhypertensive | 58 | HEMODIALYSIS | 194 | Heart failure | 5 | Diseases and Disorders of the Circulatory System | 3 | Major | Major | Medical | Medicare | NA | NA | 0 | N | Y | 12310.76 | 6034.20 | Hospital | Not for Profit Corporation | 2011.4000 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 148 | F | White | Not Span/Hispanic | 9 | Emergency | 2 | Septicemia (except in labor) | 222 | BLOOD TRANSFUSION | 720 | Septicemia & disseminated infections | 18 | Infectious and Parasitic Diseases, Systemic or Unspecified Sites | 4 | Extreme | Major | Medical | Medicare | NA | NA | 0 | N | Y | 48242.20 | 20494.93 | Hospital | Not for Profit Corporation | 2277.2144 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | OOS | F | White | Not Span/Hispanic | 2 | Elective | 100 | Acute myocardial infarction | 47 | DX CARDIAC CATHETERIZTN | 190 | Acute myocardial infarction | 5 | Diseases and Disorders of the Circulatory System | 2 | Moderate | Moderate | Medical | Medicaid | NA | NA | 0 | N | N | 12794.77 | 7369.54 | Hospital | Not for Profit Corporation | 3684.7700 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 4 | Emergency | 100 | Acute myocardial infarction | 0 | NO PROC | 190 | Acute myocardial infarction | 5 | Diseases and Disorders of the Circulatory System | 2 | Moderate | Moderate | Medical | Medicare | NA | NA | 0 | N | Y | 10292.68 | 4001.00 | Hospital | Not for Profit Corporation | 1000.2500 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 18 to 29 | 141 | M | White | Not Span/Hispanic | 7 | Emergency | 130 | Pleurisy; pneumothorax; pulmonary collapse | 39 | INCISION OF PLEURA | 143 | Other respiratory diagnoses except signs, symptoms & minor diagnoses | 4 | Diseases and Disorders of the Respiratory System | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 0 | N | Y | 14177.09 | 7711.28 | Hospital | Not for Profit Corporation | 1101.6114 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 141 | M | White | Not Span/Hispanic | 9 | Emergency | 660 | Alcohol-related disorders | 231 | OTHER THERAPEUTIC PRCS | 280 | Alcoholic liver disease | 7 | Diseases and Disorders of the Hepatobiliary System and Pancreas | 3 | Major | Major | Medical | Medicare | NA | NA | 0 | N | Y | 21448.73 | 9511.37 | Hospital | Not for Profit Corporation | 1056.8189 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 0 to 17 | 147 | F | White | Not Span/Hispanic | 3 | Newborn | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 3400 | N | N | 2393.10 | 1327.02 | Hospital | Not for Profit Corporation | 442.3400 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 0 to 17 | 147 | F | White | Not Span/Hispanic | 4 | Newborn | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 2 | Moderate | Minor | Medical | Medicaid | NA | NA | 2900 | N | N | 4191.00 | 2136.17 | Hospital | Not for Profit Corporation | 534.0425 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 4 | Elective | 204 | Other non-traumatic joint disorders | 152 | ARTHROPLASTY KNEE | 302 | Knee joint replacement | 8 | Diseases and Disorders of the Musculoskeletal System and Conn Tissue | 1 | Minor | Minor | Surgical | Medicare | Private Health Insurance | NA | 0 | N | N | 29514.01 | 59253.25 | Hospital | Not for Profit Corporation | 14813.3125 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 0 to 17 | 147 | F | White | Not Span/Hispanic | 1 | Newborn | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 2700 | N | N | 813.50 | 459.49 | Hospital | Not for Profit Corporation | 459.4900 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 140 | M | White | Not Span/Hispanic | 4 | Emergency | 2 | Septicemia (except in labor) | 0 | NO PROC | 720 | Septicemia & disseminated infections | 18 | Infectious and Parasitic Diseases, Systemic or Unspecified Sites | 3 | Major | Major | Medical | Medicare | Blue Cross/Blue Shield | NA | 0 | N | Y | 10114.85 | 4943.62 | Hospital | Not for Profit Corporation | 1235.9050 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 0 to 17 | 147 | M | White | Not Span/Hispanic | 2 | Newborn | 218 | Liveborn | 115 | CIRCUMCISION | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 2600 | N | N | 1510.00 | 929.25 | Hospital | Not for Profit Corporation | 464.6250 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | M | White | Not Span/Hispanic | 2 | Emergency | 129 | Aspiration pneumonitis; food/vomitus | 0 | NO PROC | 137 | Major respiratory infections & inflammations | 4 | Diseases and Disorders of the Respiratory System | 3 | Major | Moderate | Medical | Medicaid | NA | NA | 0 | N | Y | 7330.35 | 2502.54 | Hospital | Not for Profit Corporation | 1251.2700 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 0 to 17 | 147 | F | White | Not Span/Hispanic | 2 | Newborn | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 2 | Moderate | Minor | Medical | Medicaid | NA | NA | 2900 | N | N | 1528.50 | 907.69 | Hospital | Not for Profit Corporation | 453.8450 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 0 to 17 | 148 | M | White | Not Span/Hispanic | 1 | Newborn | 218 | Liveborn | 115 | CIRCUMCISION | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Medicaid | Medicaid | NA | 3200 | N | N | 851.50 | 496.77 | Hospital | Not for Profit Corporation | 496.7700 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 18 to 29 | 147 | F | White | Not Span/Hispanic | 3 | Elective | 187 | Malposition; malpresentation | 134 | CESAREAN SECTION | 540 | Cesarean delivery | 14 | Pregnancy, Childbirth and the Puerperium | 1 | Minor | Minor | Surgical | Medicaid | NA | NA | 0 | N | N | 8550.41 | 4645.93 | Hospital | Not for Profit Corporation | 1548.6433 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 10 | Emergency | 51 | Other endocrine disorders | 0 | NO PROC | 424 | Other endocrine disorders | 10 | Endocrine, Nutritional and Metabolic Diseases and Disorders | 3 | Major | Major | Medical | Medicare | NA | NA | 0 | N | Y | 17076.26 | 9286.99 | Hospital | Not for Profit Corporation | 928.6990 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 13 | Emergency | 2 | Septicemia (except in labor) | 54 | OT VASC CATH; NOT HEART | 720 | Septicemia & disseminated infections | 18 | Infectious and Parasitic Diseases, Systemic or Unspecified Sites | 4 | Extreme | Extreme | Medical | Medicare | Private Health Insurance | NA | 0 | N | Y | 58394.67 | 40505.88 | Hospital | Not for Profit Corporation | 3115.8369 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | OOS | F | White | Not Span/Hispanic | 7 | Emergency | 127 | Chronic obstructive pulmonary disease and bronchiectasis | 0 | NO PROC | 140 | Chronic obstructive pulmonary disease | 4 | Diseases and Disorders of the Respiratory System | 2 | Moderate | Moderate | Medical | Medicare | Federal/State/Local/VA | NA | 0 | N | Y | 17623.18 | 8901.52 | Hospital | Not for Profit Corporation | 1271.6457 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | M | White | Not Span/Hispanic | 2 | Emergency | 103 | Pulmonary heart disease | 0 | NO PROC | 134 | Pulmonary embolism | 4 | Diseases and Disorders of the Respiratory System | 1 | Minor | Minor | Medical | Medicaid | NA | NA | 0 | N | Y | 8016.13 | 2608.73 | Hospital | Not for Profit Corporation | 1304.3650 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 30 to 49 | 147 | F | White | Not Span/Hispanic | 7 | Emergency | 657 | Mood disorders | 0 | NO PROC | 753 | Bipolar disorders | 19 | Mental Diseases and Disorders | 2 | Moderate | Minor | Medical | Medicaid | NA | NA | 0 | N | Y | 7380.68 | 9880.01 | Hospital | Not for Profit Corporation | 1411.4300 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 0 to 17 | 147 | M | White | Not Span/Hispanic | 3 | Newborn | 218 | Liveborn | 115 | CIRCUMCISION | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 3200 | N | N | 1964.00 | 1261.62 | Hospital | Not for Profit Corporation | 420.5400 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | M | White | Not Span/Hispanic | 4 | Emergency | 11 | Cancer of head and neck | 71 | GASTROSTOMY; TEMP/PERM | 110 | Ear, nose, mouth, throat, cranial/facial malignancies | 3 | Ear, Nose, Mouth, Throat and Craniofacial Diseases and Disorders | 3 | Major | Moderate | Medical | Private Health Insurance | Federal/State/Local/VA | NA | 0 | N | Y | 7568.00 | 5248.66 | Hospital | Not for Profit Corporation | 1312.1650 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 148 | M | White | Not Span/Hispanic | 6 | Emergency | 50 | Diabetes mellitus with complications | 0 | NO PROC | 468 | Other kidney & urinary tract diagnoses, signs & symptoms | 11 | Diseases and Disorders of the Kidney and Urinary Tract | 3 | Major | Major | Medical | Medicare | Private Health Insurance | NA | 0 | N | Y | 10902.98 | 5216.32 | Hospital | Not for Profit Corporation | 869.3867 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 4 | Emergency | 127 | Chronic obstructive pulmonary disease and bronchiectasis | 216 | RESP INTUB/MECH VENTIL | 140 | Chronic obstructive pulmonary disease | 4 | Diseases and Disorders of the Respiratory System | 3 | Major | Extreme | Medical | Medicare | NA | NA | 0 | N | Y | 14544.13 | 6380.94 | Hospital | Not for Profit Corporation | 1595.2350 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 0 to 17 | 141 | F | White | Not Span/Hispanic | 2 | Newborn | 218 | Liveborn | 0 | NO PROC | 640 | Neonate birthwt >2499g, normal newborn or neonate w other problem | 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 2500 | N | N | 1216.42 | 811.52 | Hospital | Not for Profit Corporation | 405.7600 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 30 to 49 | 148 | F | White | Not Span/Hispanic | 2 | Emergency | 657 | Mood disorders | 0 | NO PROC | 753 | Bipolar disorders | 19 | Mental Diseases and Disorders | 1 | Minor | Minor | Medical | Federal/State/Local/VA | NA | NA | 0 | N | Y | 2415.05 | 1876.09 | Hospital | Not for Profit Corporation | 938.0450 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 30 to 49 | 147 | F | White | Not Span/Hispanic | 1 | Elective | 196 | Other pregnancy and delivery including normal | 137 | OT PRCS TO ASSIST DELIV | 541 | Vaginal delivery w sterilization &/or D&C | 14 | Pregnancy, Childbirth and the Puerperium | 1 | Minor | Minor | Surgical | Federal/State/Local/VA | NA | NA | 0 | N | N | 4256.44 | 3224.37 | Hospital | Not for Profit Corporation | 3224.3700 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 18 to 29 | 147 | M | White | Not Span/Hispanic | 5 | Emergency | 657 | Mood disorders | 0 | NO PROC | 751 | Major depressive disorders & other/unspecified psychoses | 19 | Mental Diseases and Disorders | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | Medicaid | NA | 0 | N | Y | 5176.78 | 6235.63 | Hospital | Not for Profit Corporation | 1247.1260 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 9 | Emergency | 42 | Secondary malignancies | 231 | OTHER THERAPEUTIC PRCS | 952 | Nonextensive procedure unrelated to principal diagnosis | 4 | Diseases and Disorders of the Respiratory System | 3 | Major | Major | Surgical | Medicare | Medicaid | NA | 0 | N | Y | 22783.03 | 12623.80 | Hospital | Not for Profit Corporation | 1402.6444 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 2 | Emergency | 127 | Chronic obstructive pulmonary disease and bronchiectasis | 0 | NO PROC | 140 | Chronic obstructive pulmonary disease | 4 | Diseases and Disorders of the Respiratory System | 2 | Moderate | Moderate | Medical | Medicare | NA | NA | 0 | N | Y | 15927.38 | 5406.24 | Hospital | Not for Profit Corporation | 2703.1200 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | M | White | Not Span/Hispanic | 3 | Emergency | 62 | Coagulation and hemorrhagic disorders | 76 | COLONOSCOPY AND BIOPSY | 661 | Coagulation & platelet disorders | 16 | Diseases and Disorders of Blood, Blood Forming Organs and Immunological Disorders | 3 | Major | Moderate | Medical | Medicare | NA | NA | 0 | N | Y | 12730.75 | 4252.01 | Hospital | Not for Profit Corporation | 1417.3367 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 6 | Emergency | 135 | Intestinal infection | 222 | BLOOD TRANSFUSION | 248 | Major gastrointestinal & peritoneal infections | 6 | Diseases and Disorders of the Digestive System | 4 | Extreme | Major | Medical | Medicare | Medicaid | NA | 0 | N | Y | 17490.31 | 9772.47 | Hospital | Not for Profit Corporation | 1628.7450 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 10 | Emergency | 157 | Acute and unspecified renal failure | 222 | BLOOD TRANSFUSION | 460 | RENAL FAILURE | 11 | Diseases and Disorders of the Kidney and Urinary Tract | 4 | Extreme | Extreme | Medical | Medicare | Private Health Insurance | NA | 0 | N | Y | 19586.28 | 8822.05 | Hospital | Not for Profit Corporation | 882.2050 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 18 to 29 | 147 | F | White | Not Span/Hispanic | 1 | Emergency | 196 | Other pregnancy and delivery including normal | 0 | NO PROC | 561 | Postpartum & post abortion diagnoses w/o procedure | 14 | Pregnancy, Childbirth and the Puerperium | 1 | Minor | Minor | Medical | Private Health Insurance | NA | NA | 0 | N | Y | 2203.28 | 894.46 | Hospital | Not for Profit Corporation | 894.4600 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 140 | F | White | Not Span/Hispanic | 6 | Emergency | 159 | Urinary tract infections | 0 | NO PROC | 463 | Kidney & urinary tract infections | 11 | Diseases and Disorders of the Kidney and Urinary Tract | 3 | Major | Major | Medical | Medicare | NA | NA | 0 | N | Y | 10387.40 | 4716.31 | Hospital | Not for Profit Corporation | 786.0517 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 2 | Elective | 44 | Neoplasms of unspecified nature or uncertain behavior | 124 | HYSTERECTOMY; AB/VAG | 513 | Uterine & adnexa procedures for non-malignancy except leiomyoma | 13 | Diseases and Disorders of the Female Reproductive System | 1 | Minor | Minor | Surgical | Blue Cross/Blue Shield | NA | NA | 0 | N | N | 15570.16 | 12042.57 | Hospital | Not for Profit Corporation | 6021.2850 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 6 | Emergency | 159 | Urinary tract infections | 54 | OT VASC CATH; NOT HEART | 463 | Kidney & urinary tract infections | 11 | Diseases and Disorders of the Kidney and Urinary Tract | 2 | Moderate | Major | Medical | Medicare | NA | NA | 0 | N | Y | 17509.05 | 9587.57 | Hospital | Not for Profit Corporation | 1597.9283 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 30 to 49 | 147 | F | White | Not Span/Hispanic | 2 | Emergency | 3 | Bacterial infection; unspecified site | 54 | OT VASC CATH; NOT HEART | 724 | Other infectious & parasitic diseases | 18 | Infectious and Parasitic Diseases, Systemic or Unspecified Sites | 3 | Major | Minor | Medical | Medicaid | NA | NA | 0 | N | Y | 12811.88 | 6991.10 | Hospital | Not for Profit Corporation | 3495.5500 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 6 | Elective | 197 | Skin and subcutaneous tissue infections | 0 | NO PROC | 383 | Cellulitis & other skin infections | 9 | Diseases and Disorders of the Skin, Subcutaneous Tissue and Breast | 2 | Moderate | Minor | Medical | Medicare | NA | NA | 0 | N | N | 10372.37 | 5325.48 | Hospital | Not for Profit Corporation | 887.5800 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 141 | F | White | Not Span/Hispanic | 3 | Emergency | 657 | Mood disorders | 0 | NO PROC | 751 | Major depressive disorders & other/unspecified psychoses | 19 | Mental Diseases and Disorders | 2 | Moderate | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 0 | N | Y | 3688.28 | 2800.36 | Hospital | Not for Profit Corporation | 933.4533 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | F | White | Not Span/Hispanic | 4 | Emergency | 153 | Gastrointestinal hemorrhage | 92 | OTHER BOWEL DX PRCS | 253 | Other & unspecified gastrointestinal hemorrhage | 6 | Diseases and Disorders of the Digestive System | 3 | Major | Extreme | Medical | Medicare | Medicaid | NA | 0 | N | Y | 11755.70 | 4987.86 | Hospital | Not for Profit Corporation | 1246.9650 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 148 | F | White | Not Span/Hispanic | 6 | Emergency | 197 | Skin and subcutaneous tissue infections | 54 | OT VASC CATH; NOT HEART | 383 | Cellulitis & other skin infections | 9 | Diseases and Disorders of the Skin, Subcutaneous Tissue and Breast | 1 | Minor | Moderate | Medical | Blue Cross/Blue Shield | NA | NA | 0 | N | Y | 14615.33 | 8883.59 | Hospital | Not for Profit Corporation | 1480.5983 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | F | White | Not Span/Hispanic | 2 | Emergency | 153 | Gastrointestinal hemorrhage | 92 | OTHER BOWEL DX PRCS | 253 | Other & unspecified gastrointestinal hemorrhage | 6 | Diseases and Disorders of the Digestive System | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 0 | N | Y | 7648.80 | 2457.26 | Hospital | Not for Profit Corporation | 1228.6300 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 18 to 29 | 147 | F | White | Not Span/Hispanic | 3 | Elective | 184 | Early or threatened labor | 137 | OT PRCS TO ASSIST DELIV | 560 | Vaginal delivery | 14 | Pregnancy, Childbirth and the Puerperium | 2 | Moderate | Minor | Medical | Medicaid | NA | NA | 0 | N | N | 5429.70 | 3396.63 | Hospital | Not for Profit Corporation | 1132.2100 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | M | White | Not Span/Hispanic | 2 | Emergency | 146 | Diverticulosis and diverticulitis | 0 | NO PROC | 244 | Diverticulitis & diverticulosis | 6 | Diseases and Disorders of the Digestive System | 2 | Moderate | Moderate | Medical | Medicare | Blue Cross/Blue Shield | Private Health Insurance | 0 | N | Y | 4622.00 | 1691.56 | Hospital | Not for Profit Corporation | 845.7800 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | OOS | M | White | Not Span/Hispanic | 2 | Elective | 100 | Acute myocardial infarction | 47 | DX CARDIAC CATHETERIZTN | 190 | Acute myocardial infarction | 5 | Diseases and Disorders of the Circulatory System | 1 | Minor | Minor | Medical | Blue Cross/Blue Shield | NA | NA | 0 | N | N | 22161.20 | 12458.00 | Hospital | Not for Profit Corporation | 6229.0000 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 50 to 69 | 147 | M | White | Not Span/Hispanic | 3 | Emergency | 226 | Fracture of neck of femur (hip) | 153 | HIP REPLACEMENT,TOT/PRT | 301 | Hip joint replacement | 8 | Diseases and Disorders of the Musculoskeletal System and Conn Tissue | 1 | Minor | Minor | Surgical | Blue Cross/Blue Shield | Private Health Insurance | NA | 0 | N | Y | 30155.82 | 55346.46 | Hospital | Not for Profit Corporation | 18448.8200 |
| 66 | Western NY | Cattaraugus | Olean General Hospital | 70 or Older | 147 | M | White | Not Span/Hispanic | 4 | Emergency | 149 | Biliary tract disease | 0 | NO PROC | 284 | Disorders of gallbladder & biliary tract | 7 | Diseases and Disorders of the Hepatobiliary System and Pancreas | 3 | Major | Major | Medical | Medicare | Medicaid | NA | 0 | N | Y | 13651.03 | 5599.37 | Hospital | Not for Profit Corporation | 1399.8425 |
The following codes are Major Diagnostic Categories (MDC) that generally correspond to a major organ system. The original 1-23 MDC codes are grouped by primary diagnosis. In 1986 MDC codes were expanded to include MDC Codes 0 (Pre-MDC), 24 (Human Immunodeficiency Virus or HIV), and 25 (Multiple Significant Trauma), the latter of which, unlike codes 1-23, are based on both the primary and secondary diagnosis. Patients are assigned Pre-MDC when their Diagnosis-Related Group (DRG) is based on a procedure, mostly transplants (heart, liver, intestinal, bone marrow, etc.) and tracheostomies in this dataset, instead of a primary diagnosis and are grouped separately due to the very resource intensive nature of these procedures which may be used to treat numerous different diagnoses across MDCs [8].
# get row of first instance of each APR.MDC.Code in order
mdc_codes <- dat[match(levels(dat$APR.MDC.Code), dat$APR.MDC.Code), c("APR.MDC.Code",
"APR.MDC.Description")]
report_table(mdc_codes, w = "650px", ec = "position: center; display: block; margin: auto;")| APR.MDC.Code | APR.MDC.Description |
|---|---|
| 0 | Pre-MDC or Ungroupable |
| 1 | Diseases and Disorders of the Nervous System |
| 2 | Diseases and Disorders of the Eye |
| 3 | Ear, Nose, Mouth, Throat and Craniofacial Diseases and Disorders |
| 4 | Diseases and Disorders of the Respiratory System |
| 5 | Diseases and Disorders of the Circulatory System |
| 6 | Diseases and Disorders of the Digestive System |
| 7 | Diseases and Disorders of the Hepatobiliary System and Pancreas |
| 8 | Diseases and Disorders of the Musculoskeletal System and Conn Tissue |
| 9 | Diseases and Disorders of the Skin, Subcutaneous Tissue and Breast |
| 10 | Endocrine, Nutritional and Metabolic Diseases and Disorders |
| 11 | Diseases and Disorders of the Kidney and Urinary Tract |
| 12 | Diseases and Disorders of the Male Reproductive System |
| 13 | Diseases and Disorders of the Female Reproductive System |
| 14 | Pregnancy, Childbirth and the Puerperium |
| 15 | Newborns and Other Neonates with Conditions Originating in the Perinatal Period |
| 16 | Diseases and Disorders of Blood, Blood Forming Organs and Immunological Disorders |
| 17 | Lymphatic, Hematopoietic, Other Malignancies, Chemotherapy and Radiotherapy |
| 18 | Infectious and Parasitic Diseases, Systemic or Unspecified Sites |
| 19 | Mental Diseases and Disorders |
| 20 | Alcohol/Drug Use and Alcohol/Drug Induced Organic Mental Disorders |
| 21 | Poisonings, Toxic Effects, Other Injuries and Other Complications of Treatment |
| 22 | Burns |
| 23 | Rehabilitation, Aftercare, Other Factors Influencing Health Status and Other Health Service Contacts |
| 24 | Human Immunodeficiency Virus Infections |
| 25 | Multiple Significant Trauma |
Length of stay goes from 1 to 121 (previously 120+) but the majority of the data lies between 1 and 30. With the most patients having stayed 2 days.
report_plot(ggplot(dat) + geom_histogram_interactive(aes(x = Length.of.Stay, tooltip = ..count..),
binwidth = 1, fill = "#69b3a2", color = "#e9ecef", alpha = 0.9) + ggtitle("Distribution of Length of Stay") +
xlab("Length of Stay") + ylab("Frequency") + scale_x_continuous(breaks = c(1,
10, 20, 30, 40), limits = c(0, 40)) + scale_y_continuous(labels = scales::comma))Median length of stay increases as age group increases. The interquartile range of length of stay also increases as age group increases.
ggplot(dat, aes(x = Age.Group, y = Length.of.Stay, fill = Age.Group)) + geom_boxplot() +
theme(legend.position = "none") + ggtitle("Length of Stay by Age Group") + xlab("Age Group") +
ylab("Length of Stay") + scale_y_continuous(breaks = c(1, 10, 20), limits = c(0,
20))The longest average length of stay among MDC groups was the 0 MDC group which corresponds to Pre-MDC or Ungroupable. As discussed earlier, the Pre-MDC group is primarily composed of transplant patients which explains the longer average time spent as an inpatient.
report_plot(ggplot(dat, aes(x = APR.MDC.Code, y = Length.of.Stay, tooltip = APR.MDC.Description)) +
geom_bar_interactive(stat = "summary", fun = "mean", fill = "#69b3a2") + ggtitle("Average Length of Stay by APR MDC Code") +
xlab("APR MDC Code") + ylab("Average Length of Stay"))As severity of illness increases, average length of stay increases. Cases labeled as extreme stay for an average of 15 days.
report_plot(ggplot(dat, aes(x = APR.Severity.of.Illness.Description, y = Length.of.Stay,
fill = APR.Severity.of.Illness.Description)) + geom_bar_interactive(stat = "summary",
fun = "mean") + theme(legend.position = "none") + ggtitle("Length of Stay by Severity of Illness") +
xlab("Severity of Illness") + ylab("Average Length of Stay"))Ownership type was one of the variables merged from the second dataset which gave additional information about facility type. There is a clear distinction in mean length of stay between the different ownership types with the longest average length of stay being for a Public Benefit Corporation (PBC). A PBC is a type of for-profit entity. Municipality owned hospitals, which are run by the local government, have the second largest average length of stay
Similarly, the boxplot shows a higher median length of stay for PBC and a larger spread.
ggplot(subset(dat, !is.na(Ownership.Type)), aes(x = reorder(Ownership.Type, Length.of.Stay),
y = Length.of.Stay, fill = Ownership.Type)) + stat_summary(fun = "mean", geom = "bar") +
theme(legend.position = "none") + ggtitle("Length of Stay by Ownership Type") +
xlab("Ownership Type") + ylab("Average Length of Stay")ggplot(dat, aes(x = Ownership.Type, y = Length.of.Stay, fill = Ownership.Type)) +
geom_boxplot() + theme(legend.position = "none") + ggtitle("Length of Stay by Ownership Type") +
xlab("Ownership Type") + ylab("Length of Stay") + scale_y_continuous(breaks = c(1,
10, 20), limits = c(0, 20))The average length of stay is greater in hospitals than in Critical Access Hospitals. Critical Access Hospitals are rural hospitals overseen by the Centers for Medicare and Medicaid Services (CMS) that meet the following criteria: 25 or fewer acute care inpatient beds, at least 35 miles from the nearest hospital, keep the average length of stay for the year at 96 hours or less for acute care patients, and have emergency care services available 24/7. By keeping their status as a Critical Access Hospital, they receive certain benefits from CMS which would Critical Access Hopsitals to keep the length of stay lower in order to keep their status. [9].
ggplot(dat, aes(x = Facility.Type, y = Length.of.Stay, fill = Facility.Type)) + stat_summary(fun = "mean",
geom = "bar") + theme(legend.position = "none") + ggtitle("Average Length of stay by Facility Type") +
xlab("Facility Type") + ylab("Average Length of Stay")Cost per day is highest in patients with elective procedures and cases of trauma.
ggplot(dat, aes(x = Type.of.Admission, y = Cost.per.Day, fill = Type.of.Admission)) +
stat_summary(fun = "mean", geom = "bar") + theme(legend.position = "none") +
ggtitle("Cost per Day by Type of Admission") + xlab("Admission Type") + ylab("Average Cost per Day")Average cost per day increased as age group increased with the highest average cost in the age group from 50 to 69. Surprisingly, average cost for those 70 or older was less than both age groups of 30 to 49 and 50 to 69.
ggplot(dat, aes(x = Age.Group, y = Cost.per.Day, fill = Age.Group)) + stat_summary(fun = "mean",
geom = "bar") + theme(legend.position = "none") + ggtitle("Cost per day by Age Group") +
xlab("Age Group") + ylab("Average Cost per day")Diseases and Disorders of the Male Reproductive System (MDC Code 12), Diseases and Disorders of the Musculoskeletal System and Connective Tissue (MDC Code 8), Diseases and Disorders of the Female Reproductive System (MDC Code 13) had the highest average costs per day.
report_plot(ggplot(dat, aes(x = APR.MDC.Code, y = Cost.per.Day, tooltip = APR.MDC.Description)) +
geom_bar_interactive(stat = "summary", fun = "mean", fill = "#69b3a2") + ggtitle("Average Cost per day by APR MDC Code") +
xlab("APR MDC Code") + ylab("Average Cost per day"))Municipality owned hospitals’ cost per patient per day is at least $1000 more on average than the other four ownership types.
ggplot(dat, aes(x = Ownership.Type, y = Cost.per.Day, fill = Ownership.Type)) + stat_summary(fun = "mean",
geom = "bar") + theme(legend.position = "none") + ggtitle("Average Cost per day by Ownership Type") +
xlab("Ownership Type") + ylab("Average Cost per day")We chose to filter out observations that were related to pregnancy and newborn data as there are unique characteristics of their features. The birthweight feature, for example, was only reported for newborns. For all other observations, the birthweight was recorded as 0.
In the figure below, we see that over 2 million observations have birth weight equal to 0.
# report_plot( ggplot(dat, aes(x = Birth.Weight)) + geom_histogram_interactive(
# aes(tooltip = ..count..), fill = '#69b3a2', alpha = 0.9) +
# ggtitle('Distribution of Birth Weight') + xlab('Birth Weight (grams)') +
# ylab('Frequency'))
report_plot(ggplot(dat) + geom_histogram_interactive(aes(x = Birth.Weight, tooltip = ..count..),
binwidth = 1, fill = "#69b3a2", color = "#e9ecef", alpha = 0.9) + ggtitle("Distribution of Birth Weight") +
xlab("Birth Weight (grams)") + ylab("Frequency"))Due to confidentiality concerns, some columns were censored if the discharge record for that observation contained any potential indication of abortion, including but not limited to: elective abortion procedures, miscarriages, ectopic pregnancies, etc.
All observations with missing data for Facility.Id, Health Service Area, and Hospital County also had Abortion.Edit.Indicator = ‘Y’. The plot below shows the MDC codes that had observations where Abortion.Edit.Indicator was equal to ‘Y’. The majority of these observations were when APR MDC Code was equal to 14 which corresponds to the MDC ‘Pregnancy, Childbirth And Puerperium’.
report_plot(ggplot(subset(dat, dat$Abortion.Edit.Indicator == "Y"), aes(x = APR.MDC.Code,
tooltip = APR.MDC.Description)) + geom_bar_interactive(fill = "#69b3a2") + ggtitle("Frequency of Observations Related to Abortion by MDC Code") +
xlab("APR MDC Code") + ylab("Frequency") + theme(legend.position = "none") +
scale_y_continuous(labels = scales::comma))Rows meeting any of the following criteria were filtered:
# get unique codes, descriptions
aprdrg <- unique(dat[, c("APR.DRG.Code", "APR.DRG.Description")])
aprdrg <- aprdrg[order(aprdrg$APR.DRG.Code), ]
rownames(aprdrg) <- NULL
ccs <- unique(dat[, c("CCS.Diagnosis.Code", "CCS.Diagnosis.Description")])
ccs <- ccs[order(ccs$CCS.Diagnosis.Code), ]
rownames(ccs) <- NULL
# terms for filtering out OB-related
obgyn_terms <- paste0(c("abortion", "amniot", "birth", "C-section", "delivery", "fetal",
"fetopelvic", "labor", "liveborn", "malposition", "natal", "neonat", "obstetric",
"OB-related", "partum", "pregnancy", "umbilical cord complication"), collapse = "|")
exclude_terms <- paste0(c("non-obstetric", "except in labor"), collapse = "|")
filter_obgyn <- function(x) {
return(grep(exclude_terms, grep(obgyn_terms, x, ignore.case = TRUE, value = TRUE),
ignore.case = TRUE, value = TRUE, invert = TRUE))
}
# get APR.DRG.Code values to filter out
aprdrg_obgyn_codes <- aprdrg[aprdrg$APR.DRG.Description %in% filter_obgyn(aprdrg$APR.DRG.Description),
"APR.DRG.Code"]
# get CCS.Diagnosis.Code values to filter out
ccs_obgyn_codes <- ccs[ccs$CCS.Diagnosis.Description %in% filter_obgyn(ccs$CCS.Diagnosis.Description),
"CCS.Diagnosis.Code"]
# filter dat to include only non-obgyn data, non-neonate data
obgyn_rows <- with(dat, which(CCS.Diagnosis.Code %in% ccs_obgyn_codes | APR.DRG.Code %in%
aprdrg_obgyn_codes | APR.MDC.Code == 14 | APR.MDC.Code == 15 | Abortion.Edit.Indicator ==
"Y" | Type.of.Admission == "Newborn" | Birth.Weight > 0))
dat <- dat[-obgyn_rows, ]
# remove columns
dat <- dat[, which(!(colnames(dat) %in% c("Abortion.Edit.Indicator", "Birth.Weight")))]After filtering, there there are 1856727 observations remaining.
Due to the amount of computing power that it would take to analyze the whole dataset, we chose to subset the data by if any of a patient’s payment types (primary, secondary, or tertiary) were Medicare. 874,388 patients had Medicare as their primary payment type, and 55,333 additional patients had Medicare as either their secondary or tertiary payment type for a total of 929,721 patients.
# Subset to only medicare patients
medicare_rows <- with(dat, which(Payment.Typology.1 == "Medicare" | Payment.Typology.2 ==
"Medicare" | Payment.Typology.3 == "Medicare"))
dat <- dat[medicare_rows, ]
# drop unused levels
dat <- droplevels(dat)The distribution of length of stay in the subset with only patients with Medicare resembles the overall distribution, but the largest frequency of length of stay in this subset is 3 days.
report_plot(ggplot(dat) + geom_histogram_interactive(aes(x = Length.of.Stay, tooltip = after_stat(count)),
binwidth = 1, fill = "#69b3a2", color = "#e9ecef", alpha = 0.9) + ggtitle("Distribution of Length of Stay") +
xlab("Length of Stay") + ylab("Frequency") + scale_x_continuous(breaks = c(1,
10, 20, 30, 40), limits = c(0, 40)) + scale_y_continuous(labels = scales::comma))The greatest number of medicare patients were admitted due to Diseases and Disorders of the Circulatory System (MDC Code 5) then Diseases and Disorders of the Respiratory System (MDC Code 4) and Diseases and Disorders of the Musculoskeletal System and Connective Tissue (MDC Code 8).
report_plot(ggplot(dat, aes(x = APR.MDC.Code, tooltip = APR.MDC.Description)) + geom_bar_interactive(fill = "#69b3a2",
color = "#e9ecef", alpha = 0.9) + ggtitle("Frequency of APR MDC Code in Medicare Patients") +
xlab("APR MDC Code") + ylab("Frequency"))As expected, the majority of patients with Medicare are 70 or older and also many in the 50 to 69 age group.
report_plot(ggplot(dat, aes(x = Age.Group)) + geom_bar_interactive(aes(tooltip = after_stat(count)),
fill = "#69b3a2", color = "#e9ecef", alpha = 0.9) + ggtitle("Frequency of Individuals with Medicare by Age Group") +
xlab("Medicare") + ylab("Frequency") + scale_y_continuous(labels = scales::comma))# Split train and test set
set.seed(123)
train_rows <- caret::createDataPartition(y = dat$Length.of.Stay, p = 0.7, list = FALSE)
train <- dat[train_rows, ]
test <- dat[-train_rows, ]We will be using the same evaluation metrics used in the literature for the assessment of similar length of stay prediction models: \(R^2\) [1], mean absolute error (MAE) [1,2], mean relative error (MRE) [2], and root mean squared error (RMSE) [2].
\[ MAE = \frac {\sum_{i = 1}^{n} |\hat y_i - y_i|}{n} \]
\[ MRE = \frac {\sum_{i = 1}^{n} (|\hat y_i - y_i|/y_i)}{n}\]
\[ RMSE = \sqrt\frac {\sum_{i = 1}^{n} (\hat y_i - y_i)^2}{n}\] where \(\hat y_i\) is the predicted value for length of stay or cost per dayand \(y_i\) is the actual value for length of stay or cost per day.
# Error functions
MAE <- function(pred, actual) {
mean(abs(actual - pred))
}
MRE <- function(pred, actual) {
mean(abs(actual - pred)/actual)
}
RMSE <- function(pred, actual) {
sqrt(mean((actual - pred)^2))
}
evaluate_model <- function(model, test, response) {
actual <- test[[response]]
pred <- predict(model, test)
# Remove rows when response == 0; MRE = infinity for those observations
if (any(test[[response]] == 0)) {
test_mre <- test[!test[[response]] == 0, ]
pred_mre <- predict(model, test_mre)
actual_mre <- test_mre[[response]]
} else {
pred_mre <- pred
actual_mre <- actual
}
return(c(`Mean Absolute Error` = round(MAE(pred, actual), digits = 2), `Mean Relative Error` = round(MRE(pred_mre,
actual_mre), digits = 2), `Root Mean Squared Error` = round(RMSE(pred, actual),
digits = 2)))
}In the following models, we consider the following variables to predict length of stay.
| Column | Description |
|---|---|
| Age.Group | Age in years at time of discharge grouped into 0 to 17, 18 to 29, 30 to 49, 50 to 69, 70+ |
| Gender | Male, Female, Unknown |
| Race | Black/African American, Multi, Other Race, Unknown, White. |
| Ethnicity | Spanish/Hispanic Origin, Not of Spanish/Hispanic Origin, Multi, Unknown. |
| Type.of.Admission | Elective, Emergency, Newborn, Not Available, Trauma, Urgent. |
| APR.MDC.Code | 0-25 |
| APR.Severity.of.Illness.Code | Minor (1), Moderate (2), Major (3) , Extreme (4). |
| APR.Risk.of.Mortality | Minor (1), Moderate (2), Major (3) , Extreme (4). |
| Ownership.Type | County, Municipality, Not for Profit Corporation, Public Benefit Corporation, State |
| Facility.Type | Hospital, Primary Care Hospital - Critical Access Hospital |
| Emergency.Department.Indicator | If the record contained an Emergency Department revenue code of 045X, the indicator is set to “Y”, otherwise it will be “N”. |
| Health.Service.Area | 8 service areas represented |
| Hospital.County | 57 Hospital counties represented |
All of these variables are recorded at admission so no future information such as Total Cost is included in the model.
We wanted to focus our efforts in predicting length of stay through regression methods and not classification methods because knowing the exact number of days a patient would need to stay would be overall more useful than knowing a range of days a patient may stay. We start with a simple linear regression as our base model although we acknowledge that fundamentally, this is not the proper model that we want to use to model length of stay.
linearReg <- lm(Length.of.Stay ~ Age.Group + Gender + Race + Ethnicity + Type.of.Admission +
APR.MDC.Code + APR.Severity.of.Illness.Code + APR.Risk.of.Mortality + Ownership.Type +
Facility.Type + Emergency.Department.Indicator + Health.Service.Area + Hospital.County,
data = train)
summary(linearReg)##
## Call:
## lm(formula = Length.of.Stay ~ Age.Group + Gender + Race + Ethnicity +
## Type.of.Admission + APR.MDC.Code + APR.Severity.of.Illness.Code +
## APR.Risk.of.Mortality + Ownership.Type + Facility.Type +
## Emergency.Department.Indicator + Health.Service.Area + Hospital.County,
## data = train)
##
## Residuals:
## Min 1Q Median 3Q Max
## -24.266 -2.921 -0.990 1.348 117.851
##
## Coefficients: (7 not defined because of singularities)
## Estimate
## (Intercept) 1.84581
## Age.Group18 to 29 2.23548
## Age.Group30 to 49 2.10698
## Age.Group50 to 69 2.22973
## Age.Group70 or Older 1.77959
## GenderM -0.06179
## GenderU -0.15782
## RaceMulti-racial -0.05776
## RaceOther Race -0.51203
## RaceWhite -0.39648
## EthnicityNot Span/Hispanic 0.27230
## EthnicitySpanish/Hispanic -0.08276
## EthnicityUnknown 0.24512
## Type.of.AdmissionEmergency -0.58307
## Type.of.AdmissionNot Available -1.01670
## Type.of.AdmissionTrauma 0.41914
## Type.of.AdmissionUrgent 0.83830
## APR.MDC.Code2 -0.79959
## APR.MDC.Code3 -0.93989
## APR.MDC.Code4 -0.63769
## APR.MDC.Code5 -1.11459
## APR.MDC.Code6 -0.08713
## APR.MDC.Code7 -0.26711
## APR.MDC.Code8 -0.06009
## APR.MDC.Code9 0.04053
## APR.MDC.Code10 -1.29699
## APR.MDC.Code11 -0.98647
## APR.MDC.Code12 -1.10269
## APR.MDC.Code13 -0.83214
## APR.MDC.Code16 -0.68514
## APR.MDC.Code17 1.97804
## APR.MDC.Code18 -0.58273
## APR.MDC.Code19 9.70821
## APR.MDC.Code20 1.90188
## APR.MDC.Code21 -1.38021
## APR.MDC.Code22 5.07095
## APR.MDC.Code23 4.26049
## APR.MDC.Code24 -0.60053
## APR.MDC.Code25 0.93427
## APR.Severity.of.Illness.Code2 1.10670
## APR.Severity.of.Illness.Code3 3.16984
## APR.Severity.of.Illness.Code4 9.06343
## APR.Risk.of.MortalityMajor -1.40676
## APR.Risk.of.MortalityMinor -3.26966
## APR.Risk.of.MortalityModerate -2.47810
## Ownership.TypeMunicipality 4.55932
## Ownership.TypeNot for Profit Corporation 3.21298
## Ownership.TypePublic Benefit Corporation 5.31600
## Ownership.TypeState 3.51789
## Facility.TypePrimary Care Hospital - Critical Access Hospital -0.54996
## Emergency.Department.IndicatorY -0.27390
## Health.Service.AreaCentral NY -1.33232
## Health.Service.AreaFinger Lakes -2.83891
## Health.Service.AreaHudson Valley 0.46858
## Health.Service.AreaLong Island 0.11763
## Health.Service.AreaNew York City 0.44096
## Health.Service.AreaSouthern Tier -1.28723
## Health.Service.AreaWestern NY 2.29715
## Hospital.CountyAllegany -3.42785
## Hospital.CountyBronx 0.11297
## Hospital.CountyBroome 0.52047
## Hospital.CountyCattaraugus -4.07397
## Hospital.CountyCayuga -0.29408
## Hospital.CountyChautauqua -3.68636
## Hospital.CountyChemung 2.02711
## Hospital.CountyChenango NA
## Hospital.CountyClinton -0.06400
## Hospital.CountyColumbia 0.22025
## Hospital.CountyCortland -0.50571
## Hospital.CountyDelaware -0.46812
## Hospital.CountyDutchess -1.61482
## Hospital.CountyErie -3.07511
## Hospital.CountyEssex -0.64575
## Hospital.CountyFranklin -1.80682
## Hospital.CountyFulton -1.38847
## Hospital.CountyGenesee -3.84128
## Hospital.CountyHerkimer 0.65630
## Hospital.CountyJefferson 1.23635
## Hospital.CountyKings -0.13454
## Hospital.CountyLewis 3.00428
## Hospital.CountyLivingston 2.78739
## Hospital.CountyMadison -0.10136
## Hospital.CountyManhattan -0.48479
## Hospital.CountyMonroe 3.01853
## Hospital.CountyMontgomery -1.25763
## Hospital.CountyNassau 0.25982
## Hospital.CountyNiagara -2.50530
## Hospital.CountyOneida 0.79021
## Hospital.CountyOnondaga 1.19327
## Hospital.CountyOntario 2.04596
## Hospital.CountyOrange -1.22661
## Hospital.CountyOrleans -1.58778
## Hospital.CountyOswego 0.10542
## Hospital.CountyOtsego -0.42158
## Hospital.CountyPutnam -2.48192
## Hospital.CountyQueens 0.11431
## Hospital.CountyRensselaer -0.23273
## Hospital.CountyRichmond NA
## Hospital.CountyRockland -0.76132
## Hospital.CountySaratoga 0.12449
## Hospital.CountySchenectady -0.28084
## Hospital.CountySchuyler 2.64624
## Hospital.CountySt Lawrence 0.09196
## Hospital.CountySteuben 1.04735
## Hospital.CountySuffolk NA
## Hospital.CountySullivan -1.35045
## Hospital.CountyTompkins NA
## Hospital.CountyUlster -0.71042
## Hospital.CountyWarren -0.45330
## Hospital.CountyWayne 1.34869
## Hospital.CountyWestchester NA
## Hospital.CountyWyoming NA
## Hospital.CountyYates NA
## Std. Error
## (Intercept) 1.79931
## Age.Group18 to 29 0.44303
## Age.Group30 to 49 0.43226
## Age.Group50 to 69 0.43063
## Age.Group70 or Older 0.43064
## GenderM 0.01868
## GenderU 7.35333
## RaceMulti-racial 0.10400
## RaceOther Race 0.03540
## RaceWhite 0.02818
## EthnicityNot Span/Hispanic 0.18919
## EthnicitySpanish/Hispanic 0.19161
## EthnicityUnknown 0.19552
## Type.of.AdmissionEmergency 0.04401
## Type.of.AdmissionNot Available 0.39477
## Type.of.AdmissionTrauma 0.21100
## Type.of.AdmissionUrgent 0.04806
## APR.MDC.Code2 0.25397
## APR.MDC.Code3 0.09109
## APR.MDC.Code4 0.04232
## APR.MDC.Code5 0.03858
## APR.MDC.Code6 0.04359
## APR.MDC.Code7 0.06509
## APR.MDC.Code8 0.04343
## APR.MDC.Code9 0.06393
## APR.MDC.Code10 0.06051
## APR.MDC.Code11 0.04776
## APR.MDC.Code12 0.13021
## APR.MDC.Code13 0.12689
## APR.MDC.Code16 0.07977
## APR.MDC.Code17 0.10052
## APR.MDC.Code18 0.04615
## APR.MDC.Code19 0.06196
## APR.MDC.Code20 0.09397
## APR.MDC.Code21 0.09700
## APR.MDC.Code22 0.48111
## APR.MDC.Code23 0.06708
## APR.MDC.Code24 0.19960
## APR.MDC.Code25 0.23877
## APR.Severity.of.Illness.Code2 0.02955
## APR.Severity.of.Illness.Code3 0.03704
## APR.Severity.of.Illness.Code4 0.05580
## APR.Risk.of.MortalityMajor 0.04547
## APR.Risk.of.MortalityMinor 0.05615
## APR.Risk.of.MortalityModerate 0.05051
## Ownership.TypeMunicipality 1.73478
## Ownership.TypeNot for Profit Corporation 1.73413
## Ownership.TypePublic Benefit Corporation 1.73583
## Ownership.TypeState 1.73500
## Facility.TypePrimary Care Hospital - Critical Access Hospital 0.24278
## Emergency.Department.IndicatorY 0.03735
## Health.Service.AreaCentral NY 0.17230
## Health.Service.AreaFinger Lakes 0.48471
## Health.Service.AreaHudson Valley 0.06724
## Health.Service.AreaLong Island 0.06306
## Health.Service.AreaNew York City 0.07773
## Health.Service.AreaSouthern Tier 0.30699
## Health.Service.AreaWestern NY 1.75606
## Hospital.CountyAllegany 1.77752
## Hospital.CountyBronx 0.06954
## Hospital.CountyBroome 0.31256
## Hospital.CountyCattaraugus 1.76296
## Hospital.CountyCayuga 0.23443
## Hospital.CountyChautauqua 1.75993
## Hospital.CountyChemung 0.49322
## Hospital.CountyChenango NA
## Hospital.CountyClinton 0.14127
## Hospital.CountyColumbia 0.16128
## Hospital.CountyCortland 0.25781
## Hospital.CountyDelaware 0.48678
## Hospital.CountyDutchess 0.08271
## Hospital.CountyErie 1.75574
## Hospital.CountyEssex 0.53186
## Hospital.CountyFranklin 0.19326
## Hospital.CountyFulton 0.23890
## Hospital.CountyGenesee 1.76667
## Hospital.CountyHerkimer 0.45624
## Hospital.CountyJefferson 0.22276
## Hospital.CountyKings 0.06617
## Hospital.CountyLewis 1.80853
## Hospital.CountyLivingston 0.55063
## Hospital.CountyMadison 0.26063
## Hospital.CountyManhattan 0.06350
## Hospital.CountyMonroe 0.48363
## Hospital.CountyMontgomery 0.15476
## Hospital.CountyNassau 0.04711
## Hospital.CountyNiagara 1.75781
## Hospital.CountyOneida 0.17969
## Hospital.CountyOnondaga 0.17234
## Hospital.CountyOntario 0.49477
## Hospital.CountyOrange 0.07868
## Hospital.CountyOrleans 1.80369
## Hospital.CountyOswego 0.24939
## Hospital.CountyOtsego 0.12090
## Hospital.CountyPutnam 0.15278
## Hospital.CountyQueens 0.06826
## Hospital.CountyRensselaer 0.12881
## Hospital.CountyRichmond NA
## Hospital.CountyRockland 0.08744
## Hospital.CountySaratoga 0.13852
## Hospital.CountySchenectady 0.10382
## Hospital.CountySchuyler 0.58526
## Hospital.CountySt Lawrence 0.21144
## Hospital.CountySteuben 0.50472
## Hospital.CountySuffolk NA
## Hospital.CountySullivan 0.20612
## Hospital.CountyTompkins NA
## Hospital.CountyUlster 0.11837
## Hospital.CountyWarren 0.11510
## Hospital.CountyWayne 0.51197
## Hospital.CountyWestchester NA
## Hospital.CountyWyoming NA
## Hospital.CountyYates NA
## t value Pr(>|t|)
## (Intercept) 1.026 0.304967
## Age.Group18 to 29 5.046 4.52e-07
## Age.Group30 to 49 4.874 1.09e-06
## Age.Group50 to 69 5.178 2.25e-07
## Age.Group70 or Older 4.132 3.59e-05
## GenderM -3.308 0.000941
## GenderU -0.021 0.982877
## RaceMulti-racial -0.555 0.578635
## RaceOther Race -14.464 < 2e-16
## RaceWhite -14.068 < 2e-16
## EthnicityNot Span/Hispanic 1.439 0.150081
## EthnicitySpanish/Hispanic -0.432 0.665798
## EthnicityUnknown 1.254 0.209967
## Type.of.AdmissionEmergency -13.249 < 2e-16
## Type.of.AdmissionNot Available -2.575 0.010011
## Type.of.AdmissionTrauma 1.986 0.046984
## Type.of.AdmissionUrgent 17.443 < 2e-16
## APR.MDC.Code2 -3.148 0.001642
## APR.MDC.Code3 -10.318 < 2e-16
## APR.MDC.Code4 -15.069 < 2e-16
## APR.MDC.Code5 -28.889 < 2e-16
## APR.MDC.Code6 -1.999 0.045647
## APR.MDC.Code7 -4.104 4.07e-05
## APR.MDC.Code8 -1.384 0.166487
## APR.MDC.Code9 0.634 0.526036
## APR.MDC.Code10 -21.436 < 2e-16
## APR.MDC.Code11 -20.655 < 2e-16
## APR.MDC.Code12 -8.468 < 2e-16
## APR.MDC.Code13 -6.558 5.46e-11
## APR.MDC.Code16 -8.588 < 2e-16
## APR.MDC.Code17 19.677 < 2e-16
## APR.MDC.Code18 -12.628 < 2e-16
## APR.MDC.Code19 156.693 < 2e-16
## APR.MDC.Code20 20.239 < 2e-16
## APR.MDC.Code21 -14.228 < 2e-16
## APR.MDC.Code22 10.540 < 2e-16
## APR.MDC.Code23 63.517 < 2e-16
## APR.MDC.Code24 -3.009 0.002624
## APR.MDC.Code25 3.913 9.12e-05
## APR.Severity.of.Illness.Code2 37.454 < 2e-16
## APR.Severity.of.Illness.Code3 85.579 < 2e-16
## APR.Severity.of.Illness.Code4 162.430 < 2e-16
## APR.Risk.of.MortalityMajor -30.941 < 2e-16
## APR.Risk.of.MortalityMinor -58.235 < 2e-16
## APR.Risk.of.MortalityModerate -49.066 < 2e-16
## Ownership.TypeMunicipality 2.628 0.008584
## Ownership.TypeNot for Profit Corporation 1.853 0.063913
## Ownership.TypePublic Benefit Corporation 3.063 0.002195
## Ownership.TypeState 2.028 0.042601
## Facility.TypePrimary Care Hospital - Critical Access Hospital -2.265 0.023497
## Emergency.Department.IndicatorY -7.334 2.24e-13
## Health.Service.AreaCentral NY -7.732 1.06e-14
## Health.Service.AreaFinger Lakes -5.857 4.72e-09
## Health.Service.AreaHudson Valley 6.968 3.21e-12
## Health.Service.AreaLong Island 1.865 0.062115
## Health.Service.AreaNew York City 5.673 1.41e-08
## Health.Service.AreaSouthern Tier -4.193 2.75e-05
## Health.Service.AreaWestern NY 1.308 0.190829
## Hospital.CountyAllegany -1.928 0.053800
## Hospital.CountyBronx 1.625 0.104256
## Hospital.CountyBroome 1.665 0.095875
## Hospital.CountyCattaraugus -2.311 0.020840
## Hospital.CountyCayuga -1.254 0.209683
## Hospital.CountyChautauqua -2.095 0.036206
## Hospital.CountyChemung 4.110 3.96e-05
## Hospital.CountyChenango NA NA
## Hospital.CountyClinton -0.453 0.650505
## Hospital.CountyColumbia 1.366 0.172050
## Hospital.CountyCortland -1.962 0.049813
## Hospital.CountyDelaware -0.962 0.336225
## Hospital.CountyDutchess -19.524 < 2e-16
## Hospital.CountyErie -1.751 0.079867
## Hospital.CountyEssex -1.214 0.224702
## Hospital.CountyFranklin -9.349 < 2e-16
## Hospital.CountyFulton -5.812 6.17e-09
## Hospital.CountyGenesee -2.174 0.029683
## Hospital.CountyHerkimer 1.439 0.150292
## Hospital.CountyJefferson 5.550 2.86e-08
## Hospital.CountyKings -2.033 0.042014
## Hospital.CountyLewis 1.661 0.096679
## Hospital.CountyLivingston 5.062 4.15e-07
## Hospital.CountyMadison -0.389 0.697362
## Hospital.CountyManhattan -7.634 2.27e-14
## Hospital.CountyMonroe 6.241 4.34e-10
## Hospital.CountyMontgomery -8.126 4.44e-16
## Hospital.CountyNassau 5.515 3.49e-08
## Hospital.CountyNiagara -1.425 0.154089
## Hospital.CountyOneida 4.398 1.09e-05
## Hospital.CountyOnondaga 6.924 4.40e-12
## Hospital.CountyOntario 4.135 3.55e-05
## Hospital.CountyOrange -15.590 < 2e-16
## Hospital.CountyOrleans -0.880 0.378701
## Hospital.CountyOswego 0.423 0.672527
## Hospital.CountyOtsego -3.487 0.000488
## Hospital.CountyPutnam -16.245 < 2e-16
## Hospital.CountyQueens 1.674 0.094042
## Hospital.CountyRensselaer -1.807 0.070791
## Hospital.CountyRichmond NA NA
## Hospital.CountyRockland -8.707 < 2e-16
## Hospital.CountySaratoga 0.899 0.368813
## Hospital.CountySchenectady -2.705 0.006829
## Hospital.CountySchuyler 4.522 6.14e-06
## Hospital.CountySt Lawrence 0.435 0.663630
## Hospital.CountySteuben 2.075 0.037977
## Hospital.CountySuffolk NA NA
## Hospital.CountySullivan -6.552 5.70e-11
## Hospital.CountyTompkins NA NA
## Hospital.CountyUlster -6.002 1.95e-09
## Hospital.CountyWarren -3.938 8.20e-05
## Hospital.CountyWayne 2.634 0.008431
## Hospital.CountyWestchester NA NA
## Hospital.CountyWyoming NA NA
## Hospital.CountyYates NA NA
##
## (Intercept)
## Age.Group18 to 29 ***
## Age.Group30 to 49 ***
## Age.Group50 to 69 ***
## Age.Group70 or Older ***
## GenderM ***
## GenderU
## RaceMulti-racial
## RaceOther Race ***
## RaceWhite ***
## EthnicityNot Span/Hispanic
## EthnicitySpanish/Hispanic
## EthnicityUnknown
## Type.of.AdmissionEmergency ***
## Type.of.AdmissionNot Available *
## Type.of.AdmissionTrauma *
## Type.of.AdmissionUrgent ***
## APR.MDC.Code2 **
## APR.MDC.Code3 ***
## APR.MDC.Code4 ***
## APR.MDC.Code5 ***
## APR.MDC.Code6 *
## APR.MDC.Code7 ***
## APR.MDC.Code8
## APR.MDC.Code9
## APR.MDC.Code10 ***
## APR.MDC.Code11 ***
## APR.MDC.Code12 ***
## APR.MDC.Code13 ***
## APR.MDC.Code16 ***
## APR.MDC.Code17 ***
## APR.MDC.Code18 ***
## APR.MDC.Code19 ***
## APR.MDC.Code20 ***
## APR.MDC.Code21 ***
## APR.MDC.Code22 ***
## APR.MDC.Code23 ***
## APR.MDC.Code24 **
## APR.MDC.Code25 ***
## APR.Severity.of.Illness.Code2 ***
## APR.Severity.of.Illness.Code3 ***
## APR.Severity.of.Illness.Code4 ***
## APR.Risk.of.MortalityMajor ***
## APR.Risk.of.MortalityMinor ***
## APR.Risk.of.MortalityModerate ***
## Ownership.TypeMunicipality **
## Ownership.TypeNot for Profit Corporation .
## Ownership.TypePublic Benefit Corporation **
## Ownership.TypeState *
## Facility.TypePrimary Care Hospital - Critical Access Hospital *
## Emergency.Department.IndicatorY ***
## Health.Service.AreaCentral NY ***
## Health.Service.AreaFinger Lakes ***
## Health.Service.AreaHudson Valley ***
## Health.Service.AreaLong Island .
## Health.Service.AreaNew York City ***
## Health.Service.AreaSouthern Tier ***
## Health.Service.AreaWestern NY
## Hospital.CountyAllegany .
## Hospital.CountyBronx
## Hospital.CountyBroome .
## Hospital.CountyCattaraugus *
## Hospital.CountyCayuga
## Hospital.CountyChautauqua *
## Hospital.CountyChemung ***
## Hospital.CountyChenango
## Hospital.CountyClinton
## Hospital.CountyColumbia
## Hospital.CountyCortland *
## Hospital.CountyDelaware
## Hospital.CountyDutchess ***
## Hospital.CountyErie .
## Hospital.CountyEssex
## Hospital.CountyFranklin ***
## Hospital.CountyFulton ***
## Hospital.CountyGenesee *
## Hospital.CountyHerkimer
## Hospital.CountyJefferson ***
## Hospital.CountyKings *
## Hospital.CountyLewis .
## Hospital.CountyLivingston ***
## Hospital.CountyMadison
## Hospital.CountyManhattan ***
## Hospital.CountyMonroe ***
## Hospital.CountyMontgomery ***
## Hospital.CountyNassau ***
## Hospital.CountyNiagara
## Hospital.CountyOneida ***
## Hospital.CountyOnondaga ***
## Hospital.CountyOntario ***
## Hospital.CountyOrange ***
## Hospital.CountyOrleans
## Hospital.CountyOswego
## Hospital.CountyOtsego ***
## Hospital.CountyPutnam ***
## Hospital.CountyQueens .
## Hospital.CountyRensselaer .
## Hospital.CountyRichmond
## Hospital.CountyRockland ***
## Hospital.CountySaratoga
## Hospital.CountySchenectady **
## Hospital.CountySchuyler ***
## Hospital.CountySt Lawrence
## Hospital.CountySteuben *
## Hospital.CountySuffolk
## Hospital.CountySullivan ***
## Hospital.CountyTompkins
## Hospital.CountyUlster ***
## Hospital.CountyWarren ***
## Hospital.CountyWayne **
## Hospital.CountyWestchester
## Hospital.CountyWyoming
## Hospital.CountyYates
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 7.353 on 646418 degrees of freedom
## (4282 observations deleted due to missingness)
## Multiple R-squared: 0.2083, Adjusted R-squared: 0.2081
## F-statistic: 1619 on 105 and 646418 DF, p-value: < 2.2e-16
R_squared = round(summary(linearReg)$r.squared, 4)
# Evaluate linear regression model
eval_LOS_linear <- evaluate_model(linearReg, test, "Length.of.Stay")## Error in model.frame.default(Terms, newdata, na.action = na.action, xlev = object$xlevels): factor Hospital.County has new levels Schoharie
Note that 7 coefficients are not defined due to singularities. The undefined factors are all from within the hospital county variable. We believe that this is due to the nested structure of the data. Hospital county can directly predict hospital service area because the hospital county variable is nested within the hospital service area. This is another reason that simple linear regression is not the best model for this data.
The \(R^2\) value for the simple linear regression on the training data is 0.2083.
When you look at thet possible values of length of stay, you see that length of stay can only be integer values greater than 1. Linear regression, on the other hand, will give you continuous values from -infinty to infinity. Therefore, our next step was to model length of stay as a Poisson regression which models count data. This still isn’t quite the right model as a normal Poisson can take on values of 0. Since this is inpatient data, there will never be a patient who has a length of stay of 0.
Poi <- glm(Length.of.Stay ~ Age.Group + Gender + Race + Ethnicity + Type.of.Admission +
APR.MDC.Code + APR.Severity.of.Illness.Code + APR.Risk.of.Mortality + Ownership.Type +
Facility.Type + Emergency.Department.Indicator + Health.Service.Area + Hospital.County,
family = Poisson(link = "log"), data = train)
summary(Poi)##
## Call:
## glm(formula = Length.of.Stay ~ Age.Group + Gender + Race + Ethnicity +
## Type.of.Admission + APR.MDC.Code + APR.Severity.of.Illness.Code +
## APR.Risk.of.Mortality + Ownership.Type + Facility.Type +
## Emergency.Department.Indicator + Health.Service.Area + Hospital.County,
## family = Poisson(link = "log"), data = train)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -9.9458 -1.3762 -0.5119 0.5400 25.1050
##
## Coefficients: (7 not defined because of singularities)
## Estimate
## (Intercept) 0.6651407
## Age.Group18 to 29 0.6118670
## Age.Group30 to 49 0.5741457
## Age.Group50 to 69 0.5620829
## Age.Group70 or Older 0.4884654
## GenderM -0.0088477
## GenderU -0.3109675
## RaceMulti-racial -0.0092244
## RaceOther Race -0.0764085
## RaceWhite -0.0581423
## EthnicityNot Span/Hispanic 0.0508672
## EthnicitySpanish/Hispanic -0.0023940
## EthnicityUnknown 0.0469941
## Type.of.AdmissionEmergency -0.1003854
## Type.of.AdmissionNot Available -0.1566012
## Type.of.AdmissionTrauma 0.0290503
## Type.of.AdmissionUrgent 0.0863633
## APR.MDC.Code2 -0.1547136
## APR.MDC.Code3 -0.1781237
## APR.MDC.Code4 -0.0929028
## APR.MDC.Code5 -0.1825256
## APR.MDC.Code6 -0.0131599
## APR.MDC.Code7 -0.0453772
## APR.MDC.Code8 0.0023971
## APR.MDC.Code9 0.0148520
## APR.MDC.Code10 -0.2263665
## APR.MDC.Code11 -0.1457051
## APR.MDC.Code12 -0.2064952
## APR.MDC.Code13 -0.1379889
## APR.MDC.Code16 -0.1155443
## APR.MDC.Code17 0.2242619
## APR.MDC.Code18 -0.0751799
## APR.MDC.Code19 1.1161450
## APR.MDC.Code20 0.3466617
## APR.MDC.Code21 -0.2318183
## APR.MDC.Code22 0.5165509
## APR.MDC.Code23 0.5112926
## APR.MDC.Code24 -0.1049417
## APR.MDC.Code25 0.0862924
## APR.Severity.of.Illness.Code2 0.2701772
## APR.Severity.of.Illness.Code3 0.6156490
## APR.Severity.of.Illness.Code4 1.1419976
## APR.Risk.of.MortalityMajor -0.1563404
## APR.Risk.of.MortalityMinor -0.4934566
## APR.Risk.of.MortalityModerate -0.3433385
## Ownership.TypeMunicipality 0.7533472
## Ownership.TypeNot for Profit Corporation 0.5662248
## Ownership.TypePublic Benefit Corporation 0.8396867
## Ownership.TypeState 0.6147820
## Facility.TypePrimary Care Hospital - Critical Access Hospital -0.2180760
## Emergency.Department.IndicatorY -0.0388687
## Health.Service.AreaCentral NY -0.2067186
## Health.Service.AreaFinger Lakes -0.4030761
## Health.Service.AreaHudson Valley 0.0634730
## Health.Service.AreaLong Island 0.0140725
## Health.Service.AreaNew York City 0.0629664
## Health.Service.AreaSouthern Tier -0.2825358
## Health.Service.AreaWestern NY 0.4263442
## Hospital.CountyAllegany -0.6375210
## Hospital.CountyBronx 0.0125592
## Hospital.CountyBroome 0.1548404
## Hospital.CountyCattaraugus -0.7198953
## Hospital.CountyCayuga -0.0593305
## Hospital.CountyChautauqua -0.6344370
## Hospital.CountyChemung 0.2790725
## Hospital.CountyChenango NA
## Hospital.CountyClinton -0.0012826
## Hospital.CountyColumbia 0.0294518
## Hospital.CountyCortland -0.1078135
## Hospital.CountyDelaware -0.0421928
## Hospital.CountyDutchess -0.2412427
## Hospital.CountyErie -0.5544911
## Hospital.CountyEssex -0.1460775
## Hospital.CountyFranklin -0.3103679
## Hospital.CountyFulton -0.2925608
## Hospital.CountyGenesee -0.7275779
## Hospital.CountyHerkimer 0.1837849
## Hospital.CountyJefferson 0.2005189
## Hospital.CountyKings -0.0214253
## Hospital.CountyLewis 0.5211828
## Hospital.CountyLivingston 0.3570684
## Hospital.CountyMadison -0.1127627
## Hospital.CountyManhattan -0.0729815
## Hospital.CountyMonroe 0.4248401
## Hospital.CountyMontgomery -0.1880599
## Hospital.CountyNassau 0.0412444
## Hospital.CountyNiagara -0.4589936
## Hospital.CountyOneida 0.1211710
## Hospital.CountyOnondaga 0.1795457
## Hospital.CountyOntario 0.2674217
## Hospital.CountyOrange -0.1827263
## Hospital.CountyOrleans -0.1872032
## Hospital.CountyOswego 0.0378860
## Hospital.CountyOtsego -0.0708645
## Hospital.CountyPutnam -0.4084571
## Hospital.CountyQueens 0.0132102
## Hospital.CountyRensselaer -0.0345036
## Hospital.CountyRichmond NA
## Hospital.CountyRockland -0.0939019
## Hospital.CountySaratoga 0.0222776
## Hospital.CountySchenectady -0.0679061
## Hospital.CountySchuyler 0.4324411
## Hospital.CountySt Lawrence 0.0073140
## Hospital.CountySteuben 0.0133972
## Hospital.CountySuffolk NA
## Hospital.CountySullivan -0.1928288
## Hospital.CountyTompkins NA
## Hospital.CountyUlster -0.1092904
## Hospital.CountyWarren -0.0708293
## Hospital.CountyWayne 0.1376048
## Hospital.CountyWestchester NA
## Hospital.CountyWyoming NA
## Hospital.CountyYates NA
## Std. Error
## (Intercept) 0.1258292
## Age.Group18 to 29 0.0349565
## Age.Group30 to 49 0.0346689
## Age.Group50 to 69 0.0346190
## Age.Group70 or Older 0.0346196
## GenderM 0.0009856
## GenderU 0.7071186
## RaceMulti-racial 0.0054169
## RaceOther Race 0.0018322
## RaceWhite 0.0014351
## EthnicityNot Span/Hispanic 0.0104056
## EthnicitySpanish/Hispanic 0.0105417
## EthnicityUnknown 0.0107244
## Type.of.AdmissionEmergency 0.0022921
## Type.of.AdmissionNot Available 0.0214197
## Type.of.AdmissionTrauma 0.0102718
## Type.of.AdmissionUrgent 0.0023825
## APR.MDC.Code2 0.0163307
## APR.MDC.Code3 0.0056649
## APR.MDC.Code4 0.0022750
## APR.MDC.Code5 0.0021249
## APR.MDC.Code6 0.0023771
## APR.MDC.Code7 0.0034858
## APR.MDC.Code8 0.0024334
## APR.MDC.Code9 0.0035936
## APR.MDC.Code10 0.0035903
## APR.MDC.Code11 0.0026115
## APR.MDC.Code12 0.0084623
## APR.MDC.Code13 0.0076654
## APR.MDC.Code16 0.0046056
## APR.MDC.Code17 0.0044971
## APR.MDC.Code18 0.0022854
## APR.MDC.Code19 0.0026791
## APR.MDC.Code20 0.0048427
## APR.MDC.Code21 0.0056109
## APR.MDC.Code22 0.0185804
## APR.MDC.Code23 0.0030498
## APR.MDC.Code24 0.0092524
## APR.MDC.Code25 0.0105585
## APR.Severity.of.Illness.Code2 0.0019912
## APR.Severity.of.Illness.Code3 0.0023041
## APR.Severity.of.Illness.Code4 0.0028167
## APR.Risk.of.MortalityMajor 0.0018646
## APR.Risk.of.MortalityMinor 0.0026987
## APR.Risk.of.MortalityModerate 0.0022383
## Ownership.TypeMunicipality 0.1204452
## Ownership.TypeNot for Profit Corporation 0.1204228
## Ownership.TypePublic Benefit Corporation 0.1204752
## Ownership.TypeState 0.1204572
## Facility.TypePrimary Care Hospital - Critical Access Hospital 0.0179634
## Emergency.Department.IndicatorY 0.0018928
## Health.Service.AreaCentral NY 0.0099324
## Health.Service.AreaFinger Lakes 0.0342516
## Health.Service.AreaHudson Valley 0.0035078
## Health.Service.AreaLong Island 0.0033627
## Health.Service.AreaNew York City 0.0040771
## Health.Service.AreaSouthern Tier 0.0205314
## Health.Service.AreaWestern NY 0.1215068
## Hospital.CountyAllegany 0.1227225
## Hospital.CountyBronx 0.0036041
## Hospital.CountyBroome 0.0207915
## Hospital.CountyCattaraugus 0.1218789
## Hospital.CountyCayuga 0.0138656
## Hospital.CountyChautauqua 0.1216989
## Hospital.CountyChemung 0.0346112
## Hospital.CountyChenango NA
## Hospital.CountyClinton 0.0076104
## Hospital.CountyColumbia 0.0089202
## Hospital.CountyCortland 0.0155548
## Hospital.CountyDelaware 0.0352852
## Hospital.CountyDutchess 0.0045095
## Hospital.CountyErie 0.1214956
## Hospital.CountyEssex 0.0411813
## Hospital.CountyFranklin 0.0118486
## Hospital.CountyFulton 0.0156245
## Hospital.CountyGenesee 0.1221492
## Hospital.CountyHerkimer 0.0308655
## Hospital.CountyJefferson 0.0124533
## Hospital.CountyKings 0.0034264
## Hospital.CountyLewis 0.1258666
## Hospital.CountyLivingston 0.0382218
## Hospital.CountyMadison 0.0170772
## Hospital.CountyManhattan 0.0033067
## Hospital.CountyMonroe 0.0342049
## Hospital.CountyMontgomery 0.0087912
## Hospital.CountyNassau 0.0025210
## Hospital.CountyNiagara 0.1215929
## Hospital.CountyOneida 0.0103898
## Hospital.CountyOnondaga 0.0099498
## Hospital.CountyOntario 0.0347781
## Hospital.CountyOrange 0.0042624
## Hospital.CountyOrleans 0.1241328
## Hospital.CountyOswego 0.0141012
## Hospital.CountyOtsego 0.0067461
## Hospital.CountyPutnam 0.0093391
## Hospital.CountyQueens 0.0035270
## Hospital.CountyRensselaer 0.0067658
## Hospital.CountyRichmond NA
## Hospital.CountyRockland 0.0046061
## Hospital.CountySaratoga 0.0074194
## Hospital.CountySchenectady 0.0053394
## Hospital.CountySchuyler 0.0405075
## Hospital.CountySt Lawrence 0.0123773
## Hospital.CountySteuben 0.0356921
## Hospital.CountySuffolk NA
## Hospital.CountySullivan 0.0113428
## Hospital.CountyTompkins NA
## Hospital.CountyUlster 0.0065610
## Hospital.CountyWarren 0.0062599
## Hospital.CountyWayne 0.0357522
## Hospital.CountyWestchester NA
## Hospital.CountyWyoming NA
## Hospital.CountyYates NA
## z value Pr(>|z|)
## (Intercept) 5.286 1.25e-07
## Age.Group18 to 29 17.504 < 2e-16
## Age.Group30 to 49 16.561 < 2e-16
## Age.Group50 to 69 16.236 < 2e-16
## Age.Group70 or Older 14.109 < 2e-16
## GenderM -8.977 < 2e-16
## GenderU -0.440 0.660106
## RaceMulti-racial -1.703 0.088590
## RaceOther Race -41.703 < 2e-16
## RaceWhite -40.515 < 2e-16
## EthnicityNot Span/Hispanic 4.888 1.02e-06
## EthnicitySpanish/Hispanic -0.227 0.820351
## EthnicityUnknown 4.382 1.18e-05
## Type.of.AdmissionEmergency -43.797 < 2e-16
## Type.of.AdmissionNot Available -7.311 2.65e-13
## Type.of.AdmissionTrauma 2.828 0.004681
## Type.of.AdmissionUrgent 36.249 < 2e-16
## APR.MDC.Code2 -9.474 < 2e-16
## APR.MDC.Code3 -31.443 < 2e-16
## APR.MDC.Code4 -40.837 < 2e-16
## APR.MDC.Code5 -85.896 < 2e-16
## APR.MDC.Code6 -5.536 3.09e-08
## APR.MDC.Code7 -13.018 < 2e-16
## APR.MDC.Code8 0.985 0.324581
## APR.MDC.Code9 4.133 3.58e-05
## APR.MDC.Code10 -63.049 < 2e-16
## APR.MDC.Code11 -55.794 < 2e-16
## APR.MDC.Code12 -24.402 < 2e-16
## APR.MDC.Code13 -18.002 < 2e-16
## APR.MDC.Code16 -25.088 < 2e-16
## APR.MDC.Code17 49.868 < 2e-16
## APR.MDC.Code18 -32.895 < 2e-16
## APR.MDC.Code19 416.609 < 2e-16
## APR.MDC.Code20 71.585 < 2e-16
## APR.MDC.Code21 -41.316 < 2e-16
## APR.MDC.Code22 27.801 < 2e-16
## APR.MDC.Code23 167.649 < 2e-16
## APR.MDC.Code24 -11.342 < 2e-16
## APR.MDC.Code25 8.173 3.01e-16
## APR.Severity.of.Illness.Code2 135.688 < 2e-16
## APR.Severity.of.Illness.Code3 267.193 < 2e-16
## APR.Severity.of.Illness.Code4 405.432 < 2e-16
## APR.Risk.of.MortalityMajor -83.845 < 2e-16
## APR.Risk.of.MortalityMinor -182.848 < 2e-16
## APR.Risk.of.MortalityModerate -153.390 < 2e-16
## Ownership.TypeMunicipality 6.255 3.98e-10
## Ownership.TypeNot for Profit Corporation 4.702 2.58e-06
## Ownership.TypePublic Benefit Corporation 6.970 3.17e-12
## Ownership.TypeState 5.104 3.33e-07
## Facility.TypePrimary Care Hospital - Critical Access Hospital -12.140 < 2e-16
## Emergency.Department.IndicatorY -20.535 < 2e-16
## Health.Service.AreaCentral NY -20.813 < 2e-16
## Health.Service.AreaFinger Lakes -11.768 < 2e-16
## Health.Service.AreaHudson Valley 18.095 < 2e-16
## Health.Service.AreaLong Island 4.185 2.85e-05
## Health.Service.AreaNew York City 15.444 < 2e-16
## Health.Service.AreaSouthern Tier -13.761 < 2e-16
## Health.Service.AreaWestern NY 3.509 0.000450
## Hospital.CountyAllegany -5.195 2.05e-07
## Hospital.CountyBronx 3.485 0.000493
## Hospital.CountyBroome 7.447 9.53e-14
## Hospital.CountyCattaraugus -5.907 3.49e-09
## Hospital.CountyCayuga -4.279 1.88e-05
## Hospital.CountyChautauqua -5.213 1.86e-07
## Hospital.CountyChemung 8.063 7.44e-16
## Hospital.CountyChenango NA NA
## Hospital.CountyClinton -0.169 0.866166
## Hospital.CountyColumbia 3.302 0.000961
## Hospital.CountyCortland -6.931 4.17e-12
## Hospital.CountyDelaware -1.196 0.231789
## Hospital.CountyDutchess -53.497 < 2e-16
## Hospital.CountyErie -4.564 5.02e-06
## Hospital.CountyEssex -3.547 0.000389
## Hospital.CountyFranklin -26.195 < 2e-16
## Hospital.CountyFulton -18.724 < 2e-16
## Hospital.CountyGenesee -5.956 2.58e-09
## Hospital.CountyHerkimer 5.954 2.61e-09
## Hospital.CountyJefferson 16.102 < 2e-16
## Hospital.CountyKings -6.253 4.03e-10
## Hospital.CountyLewis 4.141 3.46e-05
## Hospital.CountyLivingston 9.342 < 2e-16
## Hospital.CountyMadison -6.603 4.03e-11
## Hospital.CountyManhattan -22.071 < 2e-16
## Hospital.CountyMonroe 12.420 < 2e-16
## Hospital.CountyMontgomery -21.392 < 2e-16
## Hospital.CountyNassau 16.361 < 2e-16
## Hospital.CountyNiagara -3.775 0.000160
## Hospital.CountyOneida 11.662 < 2e-16
## Hospital.CountyOnondaga 18.045 < 2e-16
## Hospital.CountyOntario 7.689 1.48e-14
## Hospital.CountyOrange -42.869 < 2e-16
## Hospital.CountyOrleans -1.508 0.131532
## Hospital.CountyOswego 2.687 0.007216
## Hospital.CountyOtsego -10.504 < 2e-16
## Hospital.CountyPutnam -43.736 < 2e-16
## Hospital.CountyQueens 3.745 0.000180
## Hospital.CountyRensselaer -5.100 3.40e-07
## Hospital.CountyRichmond NA NA
## Hospital.CountyRockland -20.386 < 2e-16
## Hospital.CountySaratoga 3.003 0.002677
## Hospital.CountySchenectady -12.718 < 2e-16
## Hospital.CountySchuyler 10.676 < 2e-16
## Hospital.CountySt Lawrence 0.591 0.554573
## Hospital.CountySteuben 0.375 0.707397
## Hospital.CountySuffolk NA NA
## Hospital.CountySullivan -17.000 < 2e-16
## Hospital.CountyTompkins NA NA
## Hospital.CountyUlster -16.658 < 2e-16
## Hospital.CountyWarren -11.315 < 2e-16
## Hospital.CountyWayne 3.849 0.000119
## Hospital.CountyWestchester NA NA
## Hospital.CountyWyoming NA NA
## Hospital.CountyYates NA NA
##
## (Intercept) ***
## Age.Group18 to 29 ***
## Age.Group30 to 49 ***
## Age.Group50 to 69 ***
## Age.Group70 or Older ***
## GenderM ***
## GenderU
## RaceMulti-racial .
## RaceOther Race ***
## RaceWhite ***
## EthnicityNot Span/Hispanic ***
## EthnicitySpanish/Hispanic
## EthnicityUnknown ***
## Type.of.AdmissionEmergency ***
## Type.of.AdmissionNot Available ***
## Type.of.AdmissionTrauma **
## Type.of.AdmissionUrgent ***
## APR.MDC.Code2 ***
## APR.MDC.Code3 ***
## APR.MDC.Code4 ***
## APR.MDC.Code5 ***
## APR.MDC.Code6 ***
## APR.MDC.Code7 ***
## APR.MDC.Code8
## APR.MDC.Code9 ***
## APR.MDC.Code10 ***
## APR.MDC.Code11 ***
## APR.MDC.Code12 ***
## APR.MDC.Code13 ***
## APR.MDC.Code16 ***
## APR.MDC.Code17 ***
## APR.MDC.Code18 ***
## APR.MDC.Code19 ***
## APR.MDC.Code20 ***
## APR.MDC.Code21 ***
## APR.MDC.Code22 ***
## APR.MDC.Code23 ***
## APR.MDC.Code24 ***
## APR.MDC.Code25 ***
## APR.Severity.of.Illness.Code2 ***
## APR.Severity.of.Illness.Code3 ***
## APR.Severity.of.Illness.Code4 ***
## APR.Risk.of.MortalityMajor ***
## APR.Risk.of.MortalityMinor ***
## APR.Risk.of.MortalityModerate ***
## Ownership.TypeMunicipality ***
## Ownership.TypeNot for Profit Corporation ***
## Ownership.TypePublic Benefit Corporation ***
## Ownership.TypeState ***
## Facility.TypePrimary Care Hospital - Critical Access Hospital ***
## Emergency.Department.IndicatorY ***
## Health.Service.AreaCentral NY ***
## Health.Service.AreaFinger Lakes ***
## Health.Service.AreaHudson Valley ***
## Health.Service.AreaLong Island ***
## Health.Service.AreaNew York City ***
## Health.Service.AreaSouthern Tier ***
## Health.Service.AreaWestern NY ***
## Hospital.CountyAllegany ***
## Hospital.CountyBronx ***
## Hospital.CountyBroome ***
## Hospital.CountyCattaraugus ***
## Hospital.CountyCayuga ***
## Hospital.CountyChautauqua ***
## Hospital.CountyChemung ***
## Hospital.CountyChenango
## Hospital.CountyClinton
## Hospital.CountyColumbia ***
## Hospital.CountyCortland ***
## Hospital.CountyDelaware
## Hospital.CountyDutchess ***
## Hospital.CountyErie ***
## Hospital.CountyEssex ***
## Hospital.CountyFranklin ***
## Hospital.CountyFulton ***
## Hospital.CountyGenesee ***
## Hospital.CountyHerkimer ***
## Hospital.CountyJefferson ***
## Hospital.CountyKings ***
## Hospital.CountyLewis ***
## Hospital.CountyLivingston ***
## Hospital.CountyMadison ***
## Hospital.CountyManhattan ***
## Hospital.CountyMonroe ***
## Hospital.CountyMontgomery ***
## Hospital.CountyNassau ***
## Hospital.CountyNiagara ***
## Hospital.CountyOneida ***
## Hospital.CountyOnondaga ***
## Hospital.CountyOntario ***
## Hospital.CountyOrange ***
## Hospital.CountyOrleans
## Hospital.CountyOswego **
## Hospital.CountyOtsego ***
## Hospital.CountyPutnam ***
## Hospital.CountyQueens ***
## Hospital.CountyRensselaer ***
## Hospital.CountyRichmond
## Hospital.CountyRockland ***
## Hospital.CountySaratoga **
## Hospital.CountySchenectady ***
## Hospital.CountySchuyler ***
## Hospital.CountySt Lawrence
## Hospital.CountySteuben
## Hospital.CountySuffolk
## Hospital.CountySullivan ***
## Hospital.CountyTompkins
## Hospital.CountyUlster ***
## Hospital.CountyWarren ***
## Hospital.CountyWayne ***
## Hospital.CountyWestchester
## Hospital.CountyWyoming
## Hospital.CountyYates
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 3769874 on 646523 degrees of freedom
## Residual deviance: 2548733 on 646418 degrees of freedom
## (4282 observations deleted due to missingness)
## AIC: 4725404
##
## Number of Fisher Scoring iterations: 5
## Error in model.frame.default(Terms, newdata, na.action = na.action, xlev = object$xlevels): factor Hospital.County has new levels Schoharie
Note, the same hospital county coefficients were not estimated in using the Poisson model either.
Next we look at the Zero-Truncated Poisson model. This is a modification of the Poisson model, and it can only take integer values from 1 to infinity. Therfore, this model is more appropriate to model length of stay because it will never take 0 values.
# GLM with zero-truncated Poisson
ZeroTruncated_Poi <- glm(Length.of.Stay ~ Age.Group + Gender + Race + Ethnicity +
Type.of.Admission + APR.MDC.Code + APR.Severity.of.Illness.Code + APR.Risk.of.Mortality +
Ownership.Type + Facility.Type + Emergency.Department.Indicator + Health.Service.Area +
Hospital.County, family = Tpoisson(link = "log"), data = train)
summary(ZeroTruncated_Poi)
# Evaluate Zero-Truncated Poisson model
eval_LOS_ZeroTruncPoi <- evaluate_model(ZeroTruncated_Poi, test, "Length.of.Stay")Lastly, we fit a hierarchical zero-truncted poisson model. If you look at the variables for Health Serivce Area and Hospital County, you can see that these are nested varaibles. For example, specific hospital counties will only occur in specific health areas. By modeling length of stay as a multi-level model, we can take into account the area-level and county-level effects on length of stay by accounting for unexplained variation in hospital service areas and hospital counties.
# GLM with zero-truncated Poisson, nesting
Hierarchical_LOS <- HLfit(Length.of.Stay ~ Age.Group + Gender + Race + Ethnicity +
Type.of.Admission + APR.MDC.Code + APR.Severity.of.Illness.Code + APR.Risk.of.Mortality +
Ownership.Type + Facility.Type + Emergency.Department.Indicator + (1 | Hospital.County/Health.Service.Area),
family = Tpoisson(link = "log"), data = train)
summary(Hierarchical_LOS, details = TRUE, max.print = 99999999)
# Evaluate Hierarchical LOS model
eval_LOS_Hier <- evaluate_model(Hierarchical_LOS, test, "Length.of.Stay")# compare <- data.frame(matrix(, nrow = 3, ncol = 5)) colnames(compare) <- c('',
# 'Linear Regression', 'Poisson', 'Zero-Truncated Poisson','Hierarchical
# Zero-Truncated Poisson')
# #Factor compare[1,1] <- 'Mean Absolute Error' compare[2,1] <- 'Mean Relative
# Error' compare[3,1] <- 'Root Mean Squared Error'
LOS_compare <- cbind(`Linear Regression` = eval_LOS_linear, `Poisson Model` = eval_LOS_Poi,
`ZTPoisson Model` = eval_LOS_ZeroTruncPoi, `Hierarchical ZTPoisson` = eval_LOS_Hier)
# #Linear Regression compare[1,2] <- eval_LOS_linear[1] compare[2,2] <-
# eval_LOS_linear[2] compare[3,2] <- eval_LOS_linear[3]
# # Poisson compare[1,3] <- eval_LOS_Poi[1] compare[2,3] <- eval_LOS_Poi[2]
# compare[3,3] <- eval_LOS_Poi[3]
# # Zero-Truncated Poisson compare[1,4] <- eval_LOS_ZeroTruncPoi[1] compare[2,4]
# <- eval_LOS_ZeroTruncPoi[2] compare[3,4] <- eval_LOS_ZeroTruncPoi[3]
# # Hierarchical Zero Truncated Poisson compare[1,5] <- eval_LOS_Hier[1]
# compare[2,5] <- eval_LOS_Hier[2] compare[3,5] <- eval_LOS_Hier[3]
# Print table
report_table(LOS_compare, rownames = TRUE)In predicting cost per day, we focused on comparing simple linear regression results to the results for a hierarchical linear model accounting for the variance in serivce area and hospital county. We predicted that we would see an improvement in performance by using a more complicated and computationally heavy approach using the hierarchical structure since it would account for the structure of the data.
# Linear regression for cost per day
linearReg_Cost <- lm(Cost.per.Day ~ Age.Group + Gender + Race + Ethnicity + Type.of.Admission +
APR.MDC.Code + APR.Severity.of.Illness.Code + APR.Risk.of.Mortality + Ownership.Type +
Facility.Type + Emergency.Department.Indicator + Health.Service.Area + Hospital.County,
data = train)
summary(linearReg_Cost)
R_squared2 = round(summary(linearReg_Cost)$r.squared, 4)
# Evaluate Linear regression cost per day model
eval_Cost_linear <- evaluate_model(linearReg_Cost, test, "Cost.per.Day")7 coefficients were not estimated due to singularities similarly to the results from the length of stay analysis.
Hierarchical_cost <- lmer(Cost.per.Day ~ Age.Group + Gender + Race + Ethnicity +
Type.of.Admission + APR.MDC.Code + APR.Severity.of.Illness.Code + APR.Risk.of.Mortality +
Ownership.Type + Facility.Type + Emergency.Department.Indicator + (1 | Hospital.County/Health.Service.Area),
data = train)
summary(Hierarchical_cost)
# Evaluation Hierarchical model for cost per day
eval_Cost_Hier <- evaluate_model(Hierarchical_cost, test, "Cost.per.Day")# compare2 <- data.frame(matrix(, nrow = 3, ncol = 3)) colnames(compare2) <-
# c('', 'Linear Regression', 'Hierarchical Linear Regression')
# #Factor compare2[1,1] <- 'Mean Absolute Error' compare2[2,1] <- 'Mean Relative
# Error' compare2[3,1] <- 'Root Mean Squared Error'
# #Linear Regression compare2[1,2] <- eval_Cost_linear[1] compare2[2,2] <-
# eval_Cost_linear[2] compare2[3,2] <- eval_Cost_linear[3]
# # Hierarchical Linear Regression compare2[1,3] <- eval_Cost_Hier[1]
# compare2[2,3] <- eval_Cost_Hier[2] compare2[3,3] <- eval_Cost_Hier[3]
CPD_compare = cbind(`Linear Regression` = eval_Cost_linear, `Hierarchical Linear Regression` = eval_Cost_Hier)
# Print table
report_table(CPD_compare, rownames = TRUE)We did not see a large increase in predictive power as we expected by using the hierarchical linear model in either of the two predictions. The evaluation metrics between the two methods are very similar indicating that there wasn’t much to gain in performance by using the more computationally intensive method.However, the reward you do get for running the hierarchical model is that it better represents the structure of the data, and it knows how to treat the nested variables. Since linear regression does not have the knowledge of the correlation between the nested variables, it ends up not being able to estimate certain coefficients and you get the warning “prediction from a rank-deficient fit may be misleading” when trying to predict using the linear regression model. Thus, it is better to take into account the hierarchical structure of the results to be able to trust the results of your predictions.
If we wanted to use a model to predict length of stay, in practice we don’t want the model to predict -1.5 because that would not be a realistic prediction. At least, by using a zero-truncated poisson for predicting length of stay, we are taking into account that the predicted value should be at least 1.
For the cost per day analysis, we believe we may not have seen much of a difference between the two techniques because there is likely not much difference in cost between hospitals, especially between hospitals in the same general area.
[1] W.M. Tierney, J.F. Fitzgerald, M.E. Miller, M.K. James, C.J. McDonald, Predicting inpatient costs with admitting clinical data, Med Care. 33 (1995) 1–14. https://doi.org/10.1097/00005650-199501000-00001.
[2] P.-F. Tsai, P.-C. Chen, Y.-Y. Chen, H.-Y. Song, H.-M. Lin, F.-M. Lin, Q.-P. Huang, Length of Hospital Stay Prediction at the Admission Stage for Cardiology Patients Using Artificial Neural Network, Journal of Healthcare Engineering. 2016 (2016) e7035463. https://doi.org/http://dx.doi.org/10.1155/2016/7035463.
[3] C. Yang, C. Delcher, E. Shenkman, S. Ranka, Machine learning approaches for predicting high cost high need patient expenditures in health care, BioMedical Engineering OnLine. 17 (2018) 131. https://doi.org/10.1186/s12938-018-0568-3.
[4] New York State Department of Health, The Statewide Planning and Research Cooperative System (SPARCS) Inpatient De-identified File, 2017. https://health.data.ny.gov/Health/Hospital-Inpatient-Discharges-SPARCS-De-Identified/82xm-y6g8 (accessed November 3, 2020).
[5] OECD, Average length of stay in hospitals, in: Health at a Glance 2011: OECD Indicators, OECD Publishing, Paris, 2011: pp. 88–89. https://www.oecd-ilibrary.org/social-issues-migration-health/health-at-a-glance-2011/average-length-of-stay-in-hospitals_health_glance-2011-33-en (accessed November 3, 2020).
[6] G.H. Robinson, L.E. Davis, R.P. Leifer, Prediction of Hospital Length of Stay, Health Serv Res. 1 (1966) 287–300. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1067345/ (accessed November 3, 2020).
[7] New York State Department of Health, Health Facility General Information, 2014. https://health.data.ny.gov/Health/Health-Facility-General-Information/vn5v-hh5r (accessed December 5, 2020).
[8] Centers for Medicare & Medicaid Services, ICD-10-CM/PCS MS-DRG v38.0 Definitions Manual, Centers for Medicare & Medicaid Services, Baltimore, MD, 2020. https://www.cms.gov/icd10m/version38-fullcode-cms/fullcode_cms/P0001.html (accessed December 5, 2020).
[9] Rural Health Information Hub, Critical Access Hospitals (CAHs), (2019). https://www.ruralhealthinfo.org/topics/critical-access-hospitals (accessed December 5, 2020).